jobd/gc.sh

   1 #!/bin/sh
   2
   3 # NOTE: additional options can be passed to git repack by specifying
   4 #       them after the project name, for example:
   5 #         gc.sh my-project -f
   6
   7 . @basedir@/shlib.sh
   8
   9 set -e
  10
  11 if [ $# -lt 1 ]; then
  12         echo "Usage: gc.sh projname [extra-repack-args]" >&2
  13         exit 1
  14 fi
  15
  16 # Includes
  17 _shlib_done=1
  18 unset GIROCCO_SUPPRESS_AUTO_GC_UPDATE
  19 . "$cfg_basedir/jobd/maintain-auto-gc-hack.sh"
  20 . "$cfg_basedir/jobd/generate-auto-gc-update.sh"
  21 GIROCCO_SUPPRESS_AUTO_GC_UPDATE=1 && export GIROCCO_SUPPRESS_AUTO_GC_UPDATE
  22
  23 # packing options
  24 packopts="--depth=50 --window=50 --window-memory=${var_window_memory:-1g}"
  25 quiet=; [ -n "$show_progress" ] || quiet=-q
  26
  27 umask 002
  28 [ "$cfg_permission_control" != "Hooks" ] || umask 000
  29 clean_git_env
  30
  31 pidactive() {
  32         if _result="$(kill -0 "$1" 2>&1)"; then
  33                 # process exists and we have permission to signal it
  34                 return 0
  35         fi
  36         case "$_result" in *"not permitted"*)
  37                 # we do not have permission to signal the process
  38                 return 0
  39         esac
  40         # process does not exist
  41         return 1
  42 }
  43
  44 createlock() {
  45         # A .lock file should only exist for much less than a second.
  46         # If we see a stale lock file (> 1h old), remove it and then,
  47         # just in case, wait 30 seconds for any process whose .lock
  48         # we might have just removed (it's racy) to finish doing what
  49         # should take much less than a second to do.
  50         _stalelock="$(find -L "$1.lock" -maxdepth 1 -mmin +60 -print 2>/dev/null)" || :
  51         if [ -n "$_stalelock" ]; then
  52                 rm -f "$_stalelock"
  53                 sleep 30
  54         fi
  55         for _try in p p n; do
  56                 if (set -C; >"$1.lock") 2>/dev/null; then
  57                         echo "$1.lock"
  58                         return 0
  59                 fi
  60                 # delay and try again
  61                 [ "$_try" != "p" ] || sleep 1
  62         done
  63         # cannot create lock file
  64         return 1
  65 }
  66
  67 # The pre-receive script creates one ref log file per push but we want them to
  68 # be coalesced into one ref log file per day.  We are guaranteed that any files
  69 # we find to coalesce are NOT currently being written to since they are always
  70 # written first as temporary files and then moved into place.  We attempt to
  71 # transfer the most recent modification time to the coalesced log file which
  72 # would step on its mod time if it were being written to directly, but if we
  73 # find per-process ref log files then it must be a push project and the only
  74 # thing that would write directly to the main per-day log file would be a
  75 # mirror project so there's actually no conflict.
  76 # Also, if the clock is wonky (or was futzed with) we may have both YYYYMMDD
  77 # and YYYYMMDD.gz present in which case combine them into YYYYMMDD
  78 coalesce_reflogs() {
  79         [ -d reflogs ] || return 0
  80         rm -f .gc_failed
  81         find -L reflogs -maxdepth 1 -type f -name "[0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9]" -print |
  82         while read -r rname; do
  83                 if [ -e "$rname.gz" ]; then
  84                         if [ -s "$rname" ]; then
  85                                 # Presumably the .gz file must have been created before the non-gz
  86                                 # file since it had to be uncompressed at some point therefore
  87                                 # we need to append the non-gz contents to it but keep the non-gz
  88                                 # contents timestamp so we rename to YYYYMMDD_ which will sort first
  89                                 # and be picked up in the next step if we are interrupted in the middle.
  90                                 # If a YYYYMMDD_ file already exists we append to it and transfer the
  91                                 # timestamp.  Finally we transfer the YYYYMMDD_ timestamp to the result
  92                                 # and remove the YYYYMMDD_ temporary file leaving the result uncompressed.
  93                                 if [ -e "${rname}_" ]; then
  94                                         cat "$rname" >>"${rname}_"
  95                                         touch -r "$rname" "${rname}_"
  96                                         rm -f "$rname"
  97                                         ! [ -e "$rname" ]
  98                                 else
  99                                         mv "$rname" "${rname}_"
 100                                 fi
 101                                 gzip -d "$rname.gz" </dev/null
 102                                 [ -e "$rname" ] && ! [ -e "$rname.gz" ]
 103                                 cat "${rname}_" >>"$rname"
 104                                 touch -r "${rname}_" "$rname"
 105                                 rm -f "${rname}_"
 106                         else
 107                                 # Just remove the empty file to resolve the problem
 108                                 rm -f "$rname"
 109                         fi
 110                 fi
 111         done
 112         find -L reflogs -maxdepth 1 -type f -name "[0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9]_*" -print | LC_ALL=C sort |
 113         while read -r rname; do
 114                 logname="${rname%%_*}"
 115                 # If someone's been futzing with the date, the file we want to
 116                 # append to could already have been compressed, so we just uncompress
 117                 # it here.  The previous block guarantees we do not have both a compressed
 118                 # and uncompressed version present at the same time.
 119                 if [ -e "$logname.gz" ]; then
 120                         gzip -d "$logname.gz" </dev/null
 121                         [ -e "$logname" ] && ! [ -e "$logname.gz" ]
 122                 fi
 123                 cat "$rname" >>"$logname"
 124                 touch -r "$rname" "$logname"
 125                 rm -f "$rname"
 126                 if [ -e "$rname" ]; then
 127                         >.gc_failed
 128                         echo "! [$proj] failed to remove $rname" >&2
 129                         exit 1 # will only exit subshell created by "|"
 130                 fi
 131         done
 132         ! [ -e .gc_failed ]
 133 }
 134
 135 # Remove any files in reflogs that are older than $cfg_reflogs_lifetime days
 136 prune_reflogs() {
 137         [ -d reflogs ] || return 0
 138         exp="$(( ${cfg_reflogs_lifetime:-1} * 1440 ))"
 139         [ $exp -gt 0 ] || exp=1440
 140         [ $exp -le 43200 ] || exp=43200
 141         find -L reflogs -maxdepth 1 -type f -mmin "+$exp" -exec rm -f '{}' + || :
 142 }
 143
 144 # Compact any reflogs that are not today's UTC date unless a .gz version exists
 145 compact_reflogs() {
 146         [ -d reflogs ] || return 0
 147         _td="reflogs/$(TZ=UTC date '+%Y%m%d')"
 148         find -L reflogs -maxdepth 1 -type f -name "[0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9]" -print |
 149         while read -r rname; do
 150                 [ "$rname" != "$_td" ] || continue
 151                 ! [ -e "$rname.gz" ] || continue
 152                 gzip -9 "$rname" </dev/null
 153         done
 154 }
 155
 156 # return true if there's more than one objects/pack-<sha>.pack file or
 157 # ANY sha-1 files in objects
 158 is_dirty() {
 159         _packs=$(find -L objects/pack -type f -name "pack-$octet20*.pack" -print | head -n 2 | LC_ALL=C wc -l)
 160         if [ $_packs != 1 ] && [ $_packs != 0 ]; then
 161                 return 0
 162         fi
 163         _objs=$(find -L objects/$octet -type f -name "$octet19*" -print 2>/dev/null | head -n 1 | LC_ALL=C wc -l)
 164         [ $_objs -ne 0 ]
 165 }
 166
 167 # make sure combine-packs uses the correct Git executable
 168 run_combine_packs() {
 169         PATH="$var_git_exec_path:$cfg_basedir/bin:$PATH" @basedir@/jobd/combine-packs.sh "$@"
 170 }
 171
 172 # rename_pack oldnamepath newnamepath
 173 # note that .keep and .bndl files are left untouched and not moved at all!
 174 rename_pack() {
 175         [ $# -eq 2 ] && [ "$1" != "$2" ] || {
 176                 echo >&2 "[$proj] incorrect use of rename_pack function"
 177                 exit 1
 178         }
 179         # Git assumes that if the destination of the rename already exists
 180         # that it is, in fact, a copy of the same bytes so silently succeeds
 181         # without doing anything.  We duplicate that logic here.
 182         # Git checks for the .idx file first before even trying to use a pack
 183         # so it should be the last moved and the first removed.
 184         for ext in pack bitmap idx; do
 185                 [ -f "$1.$ext" ] || continue
 186                 ln "$1.$ext" "$2.$ext" >/dev/null 2>&1 ||
 187                 [ -f "$2.$ext" ] || {
 188                         echo >&2 "[$proj] unable to move $1.$ext to $2.$ext"
 189                         exit 1
 190                 }
 191         done
 192         for ext in idx pack bitmap; do
 193                 rm -f "$1.$ext"
 194         done
 195         return 0
 196 }
 197
 198 # combine the input pack(s) into a new pack (or possibly packs if packSizeLimit set)
 199 # input pack names are read from standard input one per line delimited by the first
 200 # ':', ' ' or '\n' character on the line (which allows gfi-packs to be read directly)
 201 # all arguments, if any, are passed to pack-objects as additional options
 202 # returns non-zero on failure AND creates .gc_failed in that case
 203 combine_packs() {
 204         rm -f .gc_failed
 205         find -L objects/pack -maxdepth 1 -type f -name '*.zap*' -exec rm -f '{}' + || :
 206         run_combine_packs --replace "$@" $packopts --all-progress-implied $quiet --non-empty || {
 207                 >.gc_failed
 208                 return 1
 209         }
 210         return 0
 211 }
 212
 213 # if the current directory is_gfi_mirror then repack all packs listed in gfi-packs
 214 repack_gfi_packs() {
 215         [ -n "$gfi_mirror" ] || return 0
 216         [ -d objects/pack ] || { rm -f gfi-packs; return 0; }
 217         progress "~ [$proj] redeltifying poor quality git fast-import packs"
 218         combine_packs --ignore-missing --no-reuse-delta <gfi-packs
 219         rm -f gfi-packs
 220         return 0
 221 }
 222
 223 # combine small packs into larger pack(s)
 224 # we avoid any keep, bndl or bitmap packs
 225 # if the optional argument is non-empty even a single small pack will be redeltad
 226 combine_small_packs() {
 227         _didprogress=
 228         _minsmallpacks=2
 229         if [ -n "$1" ] && [ -n "$noreusedeltaopt" ]; then
 230                 _minsmallpacks=1
 231         fi
 232         _lpo="--exclude-no-idx --exclude-keep --exclude-bitmap --exclude-bndl"
 233         _lpo="$_lpo --exclude-sfx _u --exclude-sfx _o"
 234         _lpo="$_lpo --quiet --object-limit $var_redelta_threshold objects/pack"
 235         while
 236                 _cnt="$(list_packs --count $_lpo)" || :
 237                 test "${_cnt:-0}" -ge $_minsmallpacks
 238         do
 239                 [ -n "$_didprogress" ] || {
 240                         progress "~ [$proj] combining small packs into a single larger pack"
 241                         _didprogress=1
 242                 }
 243                 _newp="$(list_packs $_lpo | combine_packs --names $noreusedeltaopt)"
 244                 _newc="$(( $(echo "$_newp" | LC_ALL=C wc -w) ))"
 245                 # be paranoid and exit the loop if we haven't reduced the number of packs
 246                 [ $_newc -lt $_cnt ] || break
 247                 _minsmallpacks=2
 248         done
 249         return 0
 250 }
 251
 252 # Unfortunately git-svn lacks the ability to store newly fetched revisions as a pack.
 253 # However, the fetch code conveniently sets .svnpack just before it runs git-svn fetch
 254 # so that it's easy to find all the objects that have been fetched by git-svn and
 255 # combine them into a pack.  The --no-reuse-delta option is meaningless here since
 256 # everything to be packed is a loose object and therefore not a delta so deltification
 257 # will always take place.
 258 make_svn_pack() {
 259         [ -f .svnpack ] && [ -n "$svn_mirror" ] || return 0
 260         rm -f .svnpackgc
 261         mv -f .svnpack .svnpackgc
 262         progress "~ [$proj] combining loose git-svn objects into a pack"
 263         _newp="$(find -L objects/$octet -maxdepth 1 -type f -newer .svnpackgc -name "$octet19*" -print 2>/dev/null |
 264         LC_ALL=C awk -F / '{print $2 $3}' |
 265         run_combine_packs --objects --names $packopts --incremental --all-progress-implied $quiet --non-empty)" || {
 266                 mv -f .svnpackgc .svnpack
 267                 >.gc_failed
 268                 return 1
 269         }
 270         if [ -n "$_newp" ]; then
 271                 # remove the now-redundant loose objects -- this is always safe
 272                 # even during a concurrent push because a reprepare_packed_git
 273                 # will be triggered if an object that should be there is not
 274                 # found thereby finding it in the new pack instead
 275                 git prune-packed $quiet
 276         fi
 277         rm -f .svnpackgc
 278 }
 279
 280 # HEADSHA="$(pack_is_complete /full/path/to/some.pack /full/path/to/packed-refs "$(cat HEAD)")"
 281 pack_is_complete() {
 282         # Must have a matching .idx file and a non-empty packed-refs file
 283         [ -s "${1%.pack}.idx" ] || return 1
 284         [ -s "$2" ] || return 1
 285         _headsha=
 286         case "$3" in
 287                 $octet20*)
 288                         _headsha="$3"
 289                         ;;
 290                 "ref: refs/"?*|"ref:refs/"?*|"refs/"?*)
 291                         _headmatch="${3#ref:}"
 292                         _headmatch="${_headmatch# }"
 293                         _headmatchpat="$(echo "$_headmatch" | LC_ALL=C sed -e 's/\([.$]\)/\\\1/g')"
 294                         _headsha="$(LC_ALL=C grep -e "^$octet20$hexdig* $_headmatchpat\$" <"$2" |
 295                                     LC_ALL=C cut -d ' ' -f 1)"
 296                         case "$_headsha" in $octet20*) :;; *)
 297                                 return 1
 298                         esac
 299                         ;;
 300                 *)
 301                         # bad HEAD
 302                         return 1
 303         esac
 304         rm -rf pack_is_complete_test
 305         mkdir pack_is_complete_test
 306         mkdir pack_is_complete_test/refs
 307         mkdir pack_is_complete_test/objects
 308         mkdir pack_is_complete_test/objects/pack
 309         echo "$_headsha" >pack_is_complete_test/HEAD
 310         ln -s "$1" pack_is_complete_test/objects/pack/
 311         ln -s "${1%.pack}.idx" pack_is_complete_test/objects/pack/
 312         ln -s "$2" pack_is_complete_test/packed-refs
 313         _count="$(git --git-dir=pack_is_complete_test rev-list --count --all 2>/dev/null)" || :
 314         rm -rf pack_is_complete_test
 315         [ -n "$_count" ] || return 1
 316         [ "$_count" -gt 0 ] 2>/dev/null || return 1
 317         echo "$_headsha"
 318 }
 319
 320 # On return a "$lockf" will have been created that must be removed when gc is done
 321 lock_gc() {
 322         # be compatibile with gc.pid file from newer Git releases
 323         lockf=gc.pid
 324         hn="$(hostname)"
 325         active=
 326         if [ "$(createlock "$lockf")" ]; then
 327                 # If $lockf is:
 328                 #   1) less than 12 hours old
 329                 #   2) contains two fields (pid hostname) NO trailing NL
 330                 #   3) the hostname is different OR the pid is still alive
 331                 # then we exit as another active process is holding the lock
 332                 if [ "$(find -L "$lockf" -maxdepth 1 -mmin -720 -print 2>/dev/null)" ]; then
 333                         apid=
 334                         ahost=
 335                         read -r apid ahost ajunk <"$lockf" || :
 336                         if [ "$apid" ] && [ "$ahost" ]; then
 337                                 if [ "$ahost" != "$hn" ] || pidactive "$apid"; then
 338                                         active=1
 339                                 fi
 340                         fi
 341                 fi
 342         else
 343                 echo >&2 "[$proj] unable to create gc.pid.lock file"
 344                 exit 1
 345         fi
 346         if [ -n "$active" ]; then
 347                 rm -f "$lockf.lock"
 348                 echo >&2 "[$proj] gc already running on machine '$ahost' pid '$apid'"
 349                 exit 1
 350         fi
 351         printf "%s %s" "$$" "$hn" >"$lockf.lock"
 352         chmod 0664 "$lockf.lock"
 353         mv -f "$lockf.lock" "$lockf"
 354 }
 355
 356 # Create a repack subdirectory such that running repack in it will pack the
 357 # same things that a pack in the normal directory would except that the pack
 358 # is guaranteed to be generated in an optimized order by adding a suitable
 359 # synthesized ref in the refs/tags namespace (yes, pack-objects.c really does
 360 # behave differently depending on the contents of the refs/tags namespace).
 361 # Before calling this, pack-refs --all MUST be performed or the wrong pack
 362 # will end up being made.
 363 #
 364 # If a ref deletion is pushed after making the repack subdir but before the
 365 # the actual repack, the discarded objects will be packed -- no big deal,
 366 # they'll get discarded the next time gc runs.
 367 #
 368 # If a fast-forward ref update is pushed after making the repack subdir but
 369 # before the actual repack, it will be picked up and the new objects packed
 370 # (subject to the normal git repack race about picking such updates up).
 371 #
 372 # If a non-fast-forward ref update is pushed after making the repack subdir but
 373 # before the actual repack, it will be picked up like a fast-forward update but
 374 # the discarded objects will be included like a ref deletion (until the next
 375 # scheduled gc takes place).
 376 #
 377 # We retain a copy of the original packed-refs file as repack/packed-refs.orig
 378 # If ref deletions come in while we're repacking, the original packed-refs
 379 # file will be modified, but we'll still pack the deleted ref(s).
 380 # If the packed-refs.orig file is used to create the bundle header we avoid
 381 # a situation where the bundle contains a ref state that never actually
 382 # existed in reality (for example a new branch is pushed and then an old
 383 # branch deleted afterwards -- the deletion would show up in the bundle
 384 # because it will cause the original packed-refs file to be re-written, but
 385 # the new branch creation will not unless we do another pack-refs which might
 386 # lead to having in incomplete bundle).  Therefore we want to keep a copy of
 387 # the original packed-refs file around.  We do the same thing for HEAD.
 388 #
 389 # It's possible that the "objects" subdirectory is a symbolic link.
 390 # Git does support this.  However, during the repacking process, new packs
 391 # will be created in repack/alt/pack and then moved into objects/pack.
 392 # In order for this to work seemlessly, they must both be on the same
 393 # filesystem.  But when objects (or even objects/pack) is a symbolic link they
 394 # might not be.  For this reason a "repack" subdirectory is created under
 395 # objects/pack and the repack/alt/pack directory symbolicly linked to it.
 396 #
 397 # Git allows not just HEAD to be a symbolic-ref, but any ref anywhere in the
 398 # refs namespace.  We are concerned about ref name collisions and getting the
 399 # right tag set to get an optimal pack.  We can safely duplicate the ref space
 400 # under refs/heads, refs/notes and refs/remotes without any risk of unwanted
 401 # collisions and this will likely make over 99%+ of all symbolic refs found
 402 # in the wild work properly.  Girocco itself never creates any symbolic refs
 403 # inside the refs namespace; this is a nod to simultaneously using a Girocco
 404 # repository for other purposes.
 405 make_repack_dir() {
 406         ! [ -d repack ] || rm -rf repack
 407         ! [ -d repack ] || { echo >&2 "[$proj] cannot remove repack subdirectory"; exit 1; }
 408         [ -d objects/pack ] || mkdir -p objects/pack
 409         ! [ -d objects/pack/repack ] || rm -rf objects/pack/repack
 410         ! [ -d objects/pack/repack ] || { echo >&2 "[$proj] cannot remove objects/pack/repack subdirectory"; exit 1; }
 411         mkdir repack repack/refs repack/alt objects/pack/repack
 412         [ -d info ] || mkdir info
 413         ln -s ../config  repack/config
 414         ln -s ../info    repack/info
 415         ln -s ../objects repack/objects
 416         ln -s "$PWD/objects/pack/repack" repack/alt/pack
 417         ln -s ../../refs repack/refs/refs
 418         _lines=$(( $(LC_ALL=C wc -l <packed-refs) ))
 419         cat HEAD >repack/HEAD.orig
 420         cat packed-refs >repack/packed-refs.orig
 421         if [ $(LC_ALL=C wc -l <repack/packed-refs.orig) -ne "$_lines" ]; then
 422                 echo >&2 "[$proj] error: make_repack_dir failed original packed-refs line count sanity check"
 423                 exit 1
 424         fi
 425         # Note: Git v1.5.0 introduced the "# pack-refs with:" header line for the packed-refs file
 426         sed '/^# pack-refs/d; s, refs/, refs/!/,' <repack/packed-refs.orig >repack/packed-refs
 427         nohead=
 428         headref="$(git rev-parse --verify --quiet HEAD)" || :
 429         if [ -n "$headref" ]; then
 430                 echo "$headref refs/!=/HEAD" >>repack/packed-refs
 431                 echo "$headref refs/heads/!" >>repack/packed-refs
 432                 nohead='\, refs/heads/!$,d; '
 433                 _lines=$(( $_lines + 2 ))
 434         fi
 435         if [ $(( $(LC_ALL=C wc -l <repack/packed-refs) + 1 )) -ne "$_lines" ]; then
 436                 echo >&2 "[$proj] error: make_repack_dir failed packed-refs initial line count sanity check"
 437                 exit 1
 438         fi
 439         sed -n "$nohead"'\, refs/heads/,p; \, refs/notes/,p; \, refs/remotes/,p' <repack/packed-refs.orig >>repack/packed-refs
 440         _newlines="$(( $(LC_ALL=C wc -l <repack/packed-refs) ))"
 441         if [ $(( $_newlines + 1 )) -lt "$_lines" ]; then
 442                 echo >&2 "[$proj] error: make_repack_dir failed packed-refs extra line count sanity check"
 443                 exit 1
 444         fi
 445         _lines="$_newlines"
 446         optref="$(git rev-list -n 1 --all 2>/dev/null)" || :
 447         if [ -n "$optref" ]; then
 448                 echo "$optref refs/tags/!" >>repack/packed-refs
 449                 _lines=$(( $_lines + 1 ))
 450                 echo "$optref" >repack/HEAD
 451         else
 452                 cat HEAD >repack/HEAD
 453         fi
 454         if [ $(LC_ALL=C wc -l <repack/packed-refs) -ne "$_lines" ]; then
 455                 echo >&2 "[$proj] error: make_repack_dir failed packed-refs line count sanity check"
 456                 exit 1
 457         fi
 458 }
 459
 460 # Remove any crud that's been left behind by interrupted operations
 461 # that did not clean up after themselves
 462 remove_crud() {
 463         # Remove any existing FETCH_HEAD
 464         # There can only be a FETCH_HEAD if we've been fetching, not if we've been
 465         # receiving pushes (those never create a FETCH_HEAD).
 466         # And if we're fetching because we're a mirror, we know we're not fetching right
 467         # now since jobd.pl never runs a project's fetch simultaneously with its gc.
 468         # Therefore any existing FETCH_HEAD is junk.  And it may be many megabytes if
 469         # there were a lot of refs.
 470         rm -f FETCH_HEAD
 471
 472         # remove any existing pack_is_complete_test or repack subdirectories
 473         # If either exists when this function is called it's crud
 474         rm -rf pack_is_complete_test repack objects/pack/repack
 475
 476         # Remove any stale pack remnants that are more than an hour old.
 477         # Stale pack fragments are defined as any pack-<sha1>.ext where .ext is NOT
 478         # .pack AND the corresponding .pack DOES NOT exist.  A bunch of stale
 479         # pack-<sha1>.idx files without their corresponding .pack files are worthless
 480         # and just waste space.  Normally there shouldn't be any remnants but actually
 481         # this can happen when things are interrupted at just the wrong time.
 482         # Note that the objects/pack directory is created by git init and should
 483         # always exist.
 484         find -L objects/pack -maxdepth 1 -type f -mmin +60 -name "pack-$octet20*.?*" -print |
 485         LC_ALL=C sed -e 's/^objects\/pack\/pack-//; s/\..*$//' | LC_ALL=C sort -u |
 486         while read packsha; do
 487                 ! [ -e "objects/pack/pack-$packsha.pack" ] || continue
 488                 rm -f "objects/pack/pack-$packsha".?*
 489         done
 490
 491         # Remove any stale tmp reflogs files that are more than one hour old.
 492         # Since they are created only while the pre-receive hook is running and
 493         # all it does is process a bunch of refs passed to it on standard input
 494         # it's inconceivable that it would ever take as much as an hour to run.
 495         if [ -d reflogs ]; then
 496                 find -L reflogs -maxdepth 1 -type f -mmin +60 -name "tmp_*" -exec rm -f '{}' + || :
 497         fi
 498
 499         # Remove any stale object tmp_obj_* files that are more than 3 hours old.
 500         # Really these files should only exist very briefly so there shouldn't be any
 501         # but things happen that can end up leaving them behind.
 502         find -L objects/$octet -maxdepth 1 -type f -mmin +180 -name "tmp_obj_?*" -exec rm -f '{}' + 2>/dev/null || :
 503
 504         # Remove any stale pack .keep files that are more than 12 hours old.
 505         # We don't do anything to create any permanent pack .keep files, so they must
 506         # be remnants from some failed push or something.  Removing the .keep will
 507         # allow the pack to be properly repacked.
 508         find -L objects/pack -maxdepth 1 -type f -mmin +720 -name "pack-$octet20*.keep" -exec rm -f '{}' + || :
 509
 510         # Remove any stale tmp_pack_*, tmp_idx_*, tmp_bitmap_*, packtmp-* or .tmp-*-pack* files
 511         # that are more than 12 hours old.
 512         find -L objects/pack -maxdepth 1 -type f -mmin +720 \( \
 513                 -name "tmp_pack_?*" -o -name "tmp_idx_?*" -o -name "tmp_bitmap_?*" -o \
 514                 -name "packtmp-?*" -o -name ".tmp-?*-pack*" \
 515                 \) -exec rm -f '{}' + || :
 516
 517         # Remove any stale incoming-* object quarantine directories that are
 518         # more than 12 hours old.  These are new with Git >= 2.11.0.
 519         find -L objects -maxdepth 1 -type d -name 'incoming-?*' -mmin +720 \
 520                 -exec rm -rf '{}' + || :
 521
 522         # Remove any stale shallow_* files that are more than 12 hours old.
 523         # These can be left behind by Git >= 1.8.4.2 and < 2.0.0 when a client
 524         # requests a shallow clone.  Also discard stale .refs-temp* and
 525         # .refs-new* files at the same time.
 526         find -L . -maxdepth 1 -type f -mmin +720 \( \
 527                 -name "shallow_?*" -o -name ".refs-temp*" -o -name ".refs-new*" \
 528                 \) -exec rm -f '{}' + || :
 529
 530         # Remove any stale *.temp files in the objects area that are more than 12 hours old.
 531         # This can be stale sha1.temp, or stale *.pack.temp so we kill all stale *.temp.
 532         find -L objects -type f -mmin +720 -name "*.temp" -exec rm -f '{}' + || :
 533
 534         # Remove any stale *.lock files in the htmlcache area that might have been left
 535         # behind after an abnormal exit during an attempt to update a cached file and
 536         # are more than 1 hour old.
 537         ! [ -d htmlcache ] || find -L htmlcache -type f -mmin +60 -name "*.lock" -exec rm -f '{}' + || :
 538
 539         # Remove any stale git-svn temp files that are more than 12 hours old.
 540         # The git-svn process creates temp files with random 10 character names
 541         # in the root of $GIT_DIR.  Unfortunately they do not have a recognizable
 542         # prefix, so we just have to kill any files with a 10-character name.  We
 543         # do this only for git-svn mirrors.  All characters are chosen from
 544         # [A-Za-z0-9_] so we can at least check that and fortunately the only
 545         # collision is 'FETCH_HEAD' but that shouldn't matter.
 546         # There may also be temp files with a Git_ prefix as well.
 547         if [ -n "$svn_mirror" ]; then
 548                 _randchar='[A-Za-z0-9_]'
 549                 _randchar2="$_randchar$_randchar"
 550                 _randchar4="$_randchar2$_randchar2"
 551                 _randchar10="$_randchar4$_randchar4$_randchar2"
 552                 find -L . -maxdepth 1 -type f -mmin +720 -name "$_randchar10" -exec rm -f '{}' + || :
 553                 find -L . -maxdepth 1 -type f -mmin +720 -name "Git_*" -exec rm -f '{}' + || :
 554         fi
 555
 556         # Remove any stale fast_import_crash_<pid> files that are more than 3 days old.
 557         if [ -n "$gfi_mirror" ]; then
 558                 find -L . -maxdepth 1 -type f -mmin +4320 -name "fast_import_crash_?*" -exec rm -f '{}' + || :
 559         fi
 560
 561         # Remove any stale core or *.core or core.* files that are more than 3 days old.
 562         find -L . -maxdepth 1 -type f -mmin +4320 \( -name "core" -o -name "*.core" -o -name "core.*" \) \
 563                 -exec rm -f '{}' + || :
 564 }
 565
 566 #
 567 ## Garbage Collection Types
 568 ##
 569 ## There are two kinds of possible garbage collection (gc) operations:
 570 ##
 571 ## 1.  A normal, full gc
 572 ## 2.  A "mini" gc
 573 ##
 574 ## If the full garbage collection interval has expired (or gc has never been
 575 ## run), then a normal, full gc will take place.  Otherwise, a "mini" gc will
 576 ## take place if the file .needsgc exists.
 577 ##
 578 ## A "mini" gc is similar to "git gc --auto" in that it may not end up actually
 579 ## doing anything unless the right conditions are present so it's not a burden
 580 ## to run it often.  If the file .needsgc exists, a "mini" gc will occur at
 581 ## the next opportunity.
 582 ##
 583 ## Note, however, that the .nogc file suppresses ALL gc activity (normal or mini).
 584 #
 585
 586 proj="${1%.git}"
 587 shift
 588 cd "$cfg_reporoot/$proj.git"
 589 [ -d objects/pack ] || { rm -f gfi-packs; mkdir -p objects/pack; }
 590 mirror_url="$(get_mirror_url)" || :
 591 svn_mirror=
 592 ! is_svn_mirror_url "$mirror_url" || svn_mirror=1
 593 gfi_mirror=
 594 if [ -f gfi-packs ] && [ -s gfi-packs ] && is_gfi_mirror_url "$mirror_url"; then
 595         gfi_mirror=1
 596 fi
 597
 598 # If git config --bool --get girocco.redelta is explicitly false then automatic
 599 # redelta when there are less than $var_redelta_threshold objects will be suppressed.
 600 # On the other hand, if git config --get girocco.redelta is "always" then, on a full
 601 # gc only, for the final repack, deltas will always be recomputed.
 602 # This can be set on a per-project basis to avoid unusual pathological gc behavior.
 603 # Setting this will hurt efficiency of the affected repository.
 604 # Note that fast-import packs ALWAYS get new deltas regardless of this setting.
 605 noreusedeltaopt="--no-reuse-delta"
 606 [ "$(git config --bool --get girocco.redelta 2>/dev/null || :)" != "false" ] || noreusedeltaopt=
 607 alwaysredelta=
 608 [ "$(git config --get girocco.redelta 2>/dev/null || :)" != "always" ] || alwaysredelta=1
 609
 610 # Extract any -f or -F or --no-reuse-object or --no-reuse-delta options
 611 # to be compatible with the old and new gc.sh versions and avoid ugly argument
 612 # duplication in process lists at the same time
 613 # Any options found will override the "girocco.redelta" setting
 614 recompress=
 615 idx=$#
 616 while [ $idx -gt 0 ]; do
 617         idx=$(( $idx - 1 ))
 618         opt="$1"
 619         shift
 620         case "$opt" in
 621         -f|--no-reuse-delta)
 622                 alwaysredelta=1
 623                 continue
 624                 ;;
 625         -F|--no-reuse-object)
 626                 alwaysredelta=1
 627                 recompress=1
 628                 continue
 629                 ;;
 630         -?*)
 631                 ;;
 632         *)
 633                 printf >&2 '%s\n' "bad non-option argument: $opt"
 634                 echo >&2 "(Did you perhaps intend to use a --xxx=yyy form?)"
 635                 exit 1
 636         esac
 637         [ -z "$opt" ] || set -- "$@" "$opt"
 638 done
 639 if [ -n "$alwaysredelta" ]; then
 640         noreusedeltaopt="--no-reuse-delta"
 641         [ -z "$recompress" ] || noreusedeltaopt="--no-reuse-object"
 642 fi
 643
 644 trap 'e=$?; rm -f .gc_in_progress; if [ $e != 0 ]; then echo "gc failed dir: $PWD" >&2; fi' EXIT
 645 trap 'exit 130' INT
 646 trap 'exit 143' TERM
 647
 648 # date -R is linux-only, POSIX equivalent is '+%a, %d %b %Y %T %z'
 649 datefmt='+%a, %d %b %Y %T %z'
 650
 651 isminigc=
 652 if [ "${force_gc:-0}" = "0" ] && check_interval lastgc $cfg_min_gc_interval; then
 653         if [ -e .needsgc ]; then
 654                 isminigc=1
 655         else
 656                 progress "= [$proj] garbage check skip (last at $(config_get lastgc))"
 657                 exit 0
 658         fi
 659 fi
 660 if [ -e .nogc ]; then
 661         progress "x [$proj] garbage check disabled"
 662         exit 0
 663 fi
 664 if [ -z "$isminigc" ] && [ -e .delaygc ] && [ -e .needsgc ]; then
 665         # Eligible for a full gc but .delaygc is set so it would be skipped
 666         # However .needsgc is also set so transform it into a mini instead
 667         isminigc=1
 668         progress "~ [$proj] garbage check delayed but checking mini because .needsgc"
 669 fi
 670
 671 if [ -n "$isminigc" ]; then
 672         # Perform a "mini" gc
 673         # Note that .delaygc is ignored here as that's only intended for full gc
 674         lock_gc
 675         rm -f .allowgc .needsgc
 676         rm -f objects/pack/pack-*_r.keep
 677         remove_crud
 678         coalesce_reflogs
 679         prune_reflogs
 680         compact_reflogs
 681         maintain_auto_gc_hack
 682         generate_auto_gc_update
 683         miniactive=
 684         if [ -f .svnpack ] && [ -n "$svn_mirror" ]; then
 685                 miniactive=1
 686                 progress "+ [$proj] mini garbage check ($(date))"
 687                 make_svn_pack
 688         fi
 689         if [ -z "$cfg_delay_gfi_redelta" ] && [ -n "$gfi_mirror" ]; then
 690                 # $Girocco::Config::delay_gfi_redelta is false, force redeltification now
 691                 if [ -z "$miniactive" ]; then
 692                         miniactive=1
 693                         progress "+ [$proj] mini garbage check ($(date))"
 694                 fi
 695                 repack_gfi_packs
 696         fi
 697         # If there aren't at least 10 non-keep, non-bitmap, non-bndl packs then
 698         # don't actually process them yet
 699         lpo="--exclude-no-idx --exclude-keep --exclude-bitmap --exclude-bndl --quiet"
 700         packcnt="$(list_packs --count $lpo objects/pack)" || :
 701         if [ "${packcnt:-0}" -ge 10 ]; then
 702                 if [ -z "$miniactive" ]; then
 703                         miniactive=1
 704                         progress "+ [$proj] mini garbage check ($(date))"
 705                 fi
 706                 if [ -n "$gfi_mirror" ]; then
 707                         repack_gfi_packs
 708                         packcnt="$(list_packs --count $lpo objects/pack)" || :
 709                 fi
 710                 # if repack_gfi_packs dropped the pack count to < 10 don't combine
 711                 if [ "${packcnt:-0}" -ge 10 ]; then
 712                         combine_small_packs
 713                         packcnt="$(list_packs --count $lpo objects/pack)" || :
 714                 fi
 715                 # if we still have more than 10 packs trigger a full gc
 716                 if [ "${packcnt:-0}" -ge 10 ]; then
 717                         # We shouldn't be in a .delaygc state at this point, but if
 718                         # we are then nuke it because we really need a full gc now
 719                         rm -f .delaygc
 720                         git config --unset gitweb.lastgc
 721                         rm -f "$lockf"
 722                         git update-server-info # just in case
 723                         progress "- [$proj] mini garbage check triggering full gc too many packs ($(date))"
 724                         exit 0
 725                 fi
 726         fi
 727         rm -f "$lockf"
 728         if [ -n "$miniactive" ]; then
 729                 git update-server-info
 730                 progress "- [$proj] mini garbage check ($(date))"
 731         else
 732                 progress "= [$proj] mini garbage check nothing but crud removal to do ($(date))"
 733         fi
 734         exit 0
 735 fi
 736
 737 # Avoid unnecessary garbage collections:
 738 #   1. If lastreceive is set and is older than lastgc
 739 #   -AND-
 740 #   2. We are not a fork (! -s alternates) -OR- lastparentgc is older than lastgc
 741
 742 # If lastgc is NOT set or lastreceive is NOT set we MUST run gc
 743 # If we are a fork and lastparentgc is NOT set we MUST run gc
 744
 745 # If the repo is dirty after removing any crud we MUST run gc
 746
 747 gcstart="$(date "$datefmt")"
 748 skipgc=
 749 isfork=
 750 ! [ -s objects/info/alternates ] || isfork=1
 751 lastparentgcsecs=
 752 [ -z "$isfork" ] || lastparentgcsecs="$(config_get_date_seconds lastparentgc)" || :
 753 lastreceivesecs=
 754 if lastreceivesecs="$(config_get_date_seconds lastreceive)" &&
 755    [ "${force_gc:-0}" = "0" ] &&
 756    lastgcsecs="$(config_get_date_seconds lastgc)" &&
 757    [ $lastreceivesecs -lt $lastgcsecs ]; then
 758         # We've run gc since we last received, so maybe we can skip,
 759         # check if not fork or fork and lastparentgc < lastgc
 760         if [ -n "$isfork" ]; then
 761                 if [ -n "$lastparentgcsecs" ] &&
 762                    [ $lastparentgcsecs -lt $lastgcsecs ]; then
 763                         # We've run gc since our parent ran gc so we can skip
 764                         skipgc=1
 765                 fi
 766         else
 767                 # We don't have any alternates (we're not a forK) so we can skip
 768                 skipgc=1
 769         fi
 770 fi
 771
 772 # Prevent any other simultaneous gc operations
 773 lock_gc
 774
 775 # At this point, if .allowgc or .gc_failed exists, it's now crud to be removed
 776 rm -f .allowgc .gc_failed
 777
 778 # Ideally we would do this in post-receive, but that would mean duplicating the
 779 # logic so it's available in the chroot jail and that's highly undesirable
 780 # Instead, since the first gc will be triggered immediately following the first
 781 # push, we do the check here as it's quick and harmless if HEAD is already valid
 782 check_and_set_head || :
 783
 784 # Always get rid of crud
 785 remove_crud
 786
 787 # Always perform reflogs maintenance
 788 coalesce_reflogs
 789 prune_reflogs
 790 compact_reflogs
 791
 792 # Always maintain auto gc hack
 793 maintain_auto_gc_hack
 794 generate_auto_gc_update
 795
 796 # Run 'git svn gc' now for svn mirrors
 797 if [ -n "$svn_mirror" ]; then
 798         git svn gc || :
 799 fi
 800
 801 # Skip the actual gc if .delaygc is set
 802 if [ -e .delaygc ]; then
 803         progress "x [$proj] garbage check delayed (except for crud removal)"
 804         rm -f "$lockf"
 805         exit 0
 806 fi
 807
 808 # Do not skip gc if the repo is dirty
 809 if [ -n "$skipgc" ] && ! is_dirty; then
 810         progress "= [$proj] garbage check nothing but crud removal to do ($(date))"
 811         config_set lastgc "$gcstart"
 812         rm -f "$lockf"
 813         exit 0
 814 fi
 815
 816 bumptime=
 817 if [ -n "$isfork" ] && [ -z "$lastparentgcsecs" ]; then
 818         # set lastparentgc and then update gcstart to be at least 1 second later
 819         config_set lastparentgc "$gcstart"
 820         bumptime=1
 821 fi
 822 if [ -z "$lastreceivesecs" ]; then
 823         # set lastreceive and then update gcstart to be at least 1 second later
 824         config_set lastreceive "$gcstart"
 825         bumptime=1
 826 fi
 827 if [ -n "$bumptime" ]; then
 828         sleep 1
 829         gcstart="$(date "$datefmt")"
 830 fi
 831
 832 progress "+ [$proj] garbage check ($(date))"
 833
 834 newdeltas=
 835 [ -z "$alwaysredelta" ] || newdeltas="$noreusedeltaopt"
 836 if [ -z "$newdeltas" ] && [ -n "$gfi_mirror" ]; then
 837         if [ $(list_packs --exclude-no-idx --count objects/pack) -le \
 838              $(list_packs --exclude-no-idx --count --quiet --only gfi-packs) ]; then
 839                 # Don't bother with repack_gfi_packs since everything's being repacked
 840                 newdeltas="--no-reuse-delta"
 841         fi
 842 fi
 843 if [ -z "$newdeltas" ] && [ -n "$noreusedeltaopt" ] &&
 844    [ $(list_packs --all --exclude-no-idx --count-objects objects/pack) -le $var_redelta_threshold ]; then
 845         # There aren't enough objects to worry about so just redelta to get the best pack
 846         newdeltas="--no-reuse-delta"
 847 fi
 848 if [ -z "$newdeltas" ]; then
 849         # Since we're not going to recompute deltas overall, we need to do the
 850         # "mini" maintenance so that we can get more optimal deltas
 851         [ -z "$noreusedeltaopt" ] || make_svn_pack
 852         repack_gfi_packs
 853         force_single_pack_redelta=
 854         [ -n "$gfi_mirror" ] || [ -n "$svn_mirror" ] || force_single_pack_redelta=1
 855         [ -z "$noreusedeltaopt" ] || combine_small_packs $force_single_pack_redelta
 856 fi
 857
 858 #
 859 ## Safe Pruning In Forks
 860 ##
 861 ## We are about to perform garbage collection.  We do NOT use the "git gc" or
 862 ## the "git repack" commands directly as they do not provide enough control over
 863 ## the fine details.  However, we DO maintain a "gc.pid" file during our garbage
 864 ## collection so that a simultaneous "git gc" by an administrator will be
 865 ## blocked (and similarly we refuse to start garbage collection if we cannot
 866 ## create the "gc.pid" file).
 867 ##
 868 ## When we say "gc" in the below description we are referring to our "gc.sh"
 869 ## script, NOT the "git gc" command.
 870 ##
 871 ## If the project we are running garbage collection (gc) on has any forks we
 872 ## must be careful not to remove any objects that while no longer referenced by
 873 ## this project (the parent) are still referenced by one or more forks (the
 874 ## children) otherwise the children will become corrupt and we can't abide
 875 ## corrupt children.
 876 ##
 877 ## One way to accomplish this is to simply hard-link all currently existing
 878 ## loose objects and packs in the parent into all the children that refer to the
 879 ## parent (via a line in their objects/info/alternates file) before beginning
 880 ## the gc operation and then relying on a subsequent gc in the child to clean up
 881 ## any excess objects/packs.  We used to use this strategy but it's very
 882 ## inefficient because:
 883 ##
 884 ##   1. The disk space used by the old pack(s)/object(s) will not be reclaimed
 885 ##      until all children (and their children, if any) run gc by which time
 886 ##      it's quite possible the topmost parent will have run gc again and
 887 ##      hard-linked yet another old pack down to its children (not to mention
 888 ##      loose objects).
 889 ##
 890 ##   2. When using the "-A" option with "git repack", any new objects in the
 891 ##      parent that are not referenced by children will continually get
 892 ##      exploded out of the hard-linked pack in the children whenever the
 893 ##      children run gc.
 894 ##
 895 ##   3. To avoid suboptimal and/or unnecessarily many packs being hard-linked
 896 ##      into child forks, we must run the "mini" gc maintenance before we
 897 ##      perform the hard-linking into the children which provides yet another
 898 ##      source of inefficiency.
 899 ##
 900 ## While we were still using the "-A" option to "git repack" (that was not
 901 ## always the case) to guarantee we can access old ref values for long enough
 902 ## to send out a meaningful mail.sh notification, another, more efficient,
 903 ## option became available to prevent corruption of child forks that continue
 904 ## to refer to objects that are no longer reachable from any ref in the parent.
 905 ##
 906 ## The only things that need be copied (or hard-linked) into the child fork(s)
 907 ## are those objects that have become unreachable from any ref in the parent.
 908 ##
 909 ## When we were using the "git repack -A -d" + "git prune --expire=1.day.ago"
 910 ## technique, the only objects that could ever be removed were loose objects
 911 ## that "git prune" determined were expired.  In that case, loose objects were
 912 ## all that need be hard-linked down to child forks in order to avoid
 913 ## corruption of any child fork(s).
 914 ##
 915 ## The "git repack -A -d" + "git prune --expire=1.day.ago" + hard-linking loose
 916 ## objects to child forks technique remains fundamentally sound from the
 917 ## perspective of supporting simultaneous gc and push and keeping newly
 918 ## unreachable objects around long enough to be sure we can send out meaningful
 919 ## ref change notifications and never corrupting any child forks and never
 920 ## persisting the lifetime of large old packs containing mostly duplicate or
 921 ## unreachable objects as gc percolates through a project's entire fork tree.
 922 ##
 923 ## However, that technique suffers from one potential prodigious pitfall.
 924 ##
 925 ## Unreachable objects come flying out of their packs to splatter all over the
 926 ## objects subdirectories possibly creating a huge, inefficient mess.
 927 ##
 928 ## Often this is not an issue.  Even with a lot of rebasing going on, usually
 929 ## the only objects that will splatter are some commits, trees and the odd blob
 930 ## here and there.  Not enough to be overly concerned about.
 931 ##
 932 ## However, for the reppository that frequently experiences a lot of non-fast-
 933 ## forward updates and/or outright ref deletion, the number of objects suddenly
 934 ## popping out of their packs at "git repack -A -d" time can be overwhelming.
 935 ##
 936 ## To avoid this issue we now use a four phase pack creation strategy.
 937 ## This will result in creation of up to four packs (instead of at most one).
 938 ##
 939 ##   I. A complete pack (with bitmaps if appropriate) gets created including
 940 ##      only "reachable" objects from all refs/... refs plus HEAD.  This will
 941 ##      also serve as the virtual bundle for the repository.
 942 ##
 943 ##  II. A pack of recently-became-unreachable objects and friends is created.
 944 ##      (The "friends" are ref logs, linked working tree HEADs and indicies.)
 945 ##      Because both the pre-receive and update.sh script record all ref
 946 ##      changes we can easily choose the cut off point for "recently".
 947 ##      It is only the fact we maintain those logs in the reflogs subdirectory
 948 ##      that allows this step to be possible.
 949 ##
 950 ## III. If the repository has any forks with a non-zero length alternates file,
 951 ##      yet another pack of "--keep-unreachable" objects is generated that will
 952 ##      not actually be kept in the parent, but hard-linked into all the forks.
 953 ##
 954 ##  IV. Finally, after running "git prune-packed", any remaining loose objects
 955 ##      are migrated into a pack of their own.
 956 ##
 957 ## We then remove any non-.keep packs that existed before we started the
 958 ## process being careful to keep any same-pack pushes for the "Push Pack Redux"
 959 ## race condition (see README-GC).
 960 ##
 961 ## By using "git pack-objects" directly we are able to accomplish this with
 962 ## very little additional effort.
 963 ##
 964 ## The packs produced by (III) are treated almost like ".keep" packs by child
 965 ## forks in that the objects in them are never repacked into any other
 966 ## "--keep-unreachable" packs (but they can migrate into phase I or II packs)
 967 ## and those phase III packs are then hard-linked into any grandchild forks.
 968 ##
 969 ## This avoids the space explosion that could occur if each fork level ended
 970 ## up duplicating the "--keep-unreachable" pack space by repacking those
 971 ## objects (essentially breaking the hard-link to the single copy of those
 972 ## objects).
 973 ##
 974 ## While it is true that each level of forks could potentially add yet another
 975 ## phase III pack to be hard-linked down to its children, such packs will only
 976 ## include unreachable objects not already in any phase III packs that were
 977 ## received from the parent.
 978 ##
 979 ## The space for the phase III packs will not be reclaimed until the gc
 980 ## finishes percolating through the entire "fork tree" of a project.
 981 ##
 982 ## This is not much different than the "git repack -A -d" situation where
 983 ## all the loose objects are hard-linked down into child forks.  In that
 984 ## case forks that actually need any of those objects could gradually reduce
 985 ## the number of objects hard-linked into deeper fork levels.
 986 ##
 987 ## The difference with a phase III "--keep-unreachable" pack is that there
 988 ## cannot be any gradual reduction like that since it would require repacking
 989 ## the pack and breaking the hard-link thereby increasing storage space.  The
 990 ## storage will instead always be reclaimed all at once when all of the
 991 ## projects in the "fork tree" complete their gc.
 992 ##
 993 ## However, the belief is that the huge space win by having all the
 994 ## unreachable objects packed up together far eclipses (when many objects are
 995 ## involved, the single-pack version can end up using 1/20th or less of the
 996 ## disk space compared to having them all as loose objects) any brief minor
 997 ## space savings that might occur under the "git repack -A -d" loose object
 998 ## system prior to the gc collection completing for all the projects in the
 999 ## "fork tree".
1000 #
1001
1002 #
1003 ## utility functions
1004 #
1005
1006 make_packs_ugw() {
1007         find -L "$1" -maxdepth 1 -type f ! -perm -ug+w \
1008                 -name "pack-$octet20*.pack" -exec chmod ug+w '{}' + || :
1009 } 2>/dev/null
1010
1011 vcnt() {
1012         eval "$1="'$(( $# - 1 ))'
1013 }
1014
1015 get_index_tree() {
1016         if [ -s "$1" ]; then
1017                 GIT_INDEX_FILE="$1"
1018                 export GIT_INDEX_FILE
1019                 git write-tree 2>/dev/null || :
1020                 unset GIT_INDEX_FILE
1021         fi
1022 }
1023
1024 get_detached_head() {
1025         if [ -s "$1" ] && read -r _head <"$1" 2>/dev/null; then
1026                 case "$_head" in $octet20*)
1027                         echo "$_head"
1028                 esac
1029         fi
1030 }
1031
1032 # compute_extra_reachables
1033 # create lines suitable for a packed-refs file mentioning all the
1034 # other refs we might like to keep.
1035 # the current directory MUST be set to the repository's --git-dir
1036 # the following are included:
1037 #   * refs mentioned in reflogs/... files
1038 #   * tree(s) created from index file(s)
1039 #   * detached linked working tree heads
1040 # Resulting objects are tested for existence and uniqified then output
1041 # one per line under a refs/z* namespace
1042 compute_extra_reachables() {
1043         {
1044                 digits8='[0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9]'
1045                 find -L reflogs -mindepth 1 -maxdepth 1 -type f -name "$digits8*" -exec gzip -c -d -f '{}' + |
1046                 awk '{print $2; print $3}'
1047                 ! [ -f index ] || get_index_tree index
1048                 if [ -d worktrees ]; then
1049                         find -L worktrees -mindepth 2 -maxdepth 2 -name HEAD -type f -print |
1050                         while read -r lwth; do
1051                                 get_detached_head "$lwth"
1052                                 get_index_tree "${lwth%HEAD}index"
1053                         done
1054                 fi
1055         } | LC_ALL=C sort -u |
1056         git cat-file ${var_have_git_260:+--buffer} --batch-check"${var_have_git_185:+=%(objectname)}" |
1057         awk '!/missing/ {num++; print $1 " " "refs/" substr("zzzzzzzzzzzz", 1, length(num)) "/" num}'
1058 }
1059
1060 #
1061 ## main gc logic
1062 #
1063
1064 # Everything else is more efficient if we do this first
1065 # The "--prune" option is the default since v1.5.0 but it serves as "documentation" here
1066 git pack-refs --all --prune
1067 [ -e packed-refs ] || >>packed-refs # should never happen...
1068
1069 # If we have a logs directory or a worktrees directory expire the ref logs now
1070 # Note that Git itself does not use either --rewrite or --updateref, so neither do we
1071 ! [ -d logs ] && ! [ -d worktrees ] || eval git reflog expire --all "${quiet:+>/dev/null 2>&1}" || :
1072
1073 make_repack_dir
1074 ! [ -e .gc_failed ] || exit 1
1075 rm -f .gc_in_progress # make sure
1076 touch .gc_in_progress # it's truly fresh
1077 rm -f bundles/* objects/pack/pack-*.bndl
1078 # This is perhaps a bit aggressive in that if we're suffering from "Push Pack Redux"
1079 # and somehow we get run again immediately after the run where "Push Pack Redux" happened
1080 # and we have garbage collection forced, there's just the barest, almost negligible,
1081 # possibility that the "Push Pack Redux" ref updates _still_ have not happened and we
1082 # should not be removing _r .keep files.  None of the normal Girocco processing can
1083 # cause this.  The second run of this script would have to use the force gc option
1084 # for it to even be possible in the first place.  What's much more likely is that
1085 # the initial run of this script was somehow interrupted in the middle before it
1086 # could get rid of the _r .keep file itself in which case it's better to get rid of
1087 # it now to avoid keeping something around that would perturb our nice and neat gc
1088 rm -f objects/pack/pack-*_r.keep
1089 # We will add .keep files for _u packs if and when we run phase III
1090 # Otherwise they need to not have any .keep files during phases I and II
1091 rm -f objects/pack/pack-*_u.keep
1092
1093 # We need to make sure that any non-Girocco (barely tolerated) Git object creation
1094 # activity will be able to "freshen" the pack containing a pre-existing object
1095 # that's being written.  This really should not be necessary as the pre-receive
1096 # hook should make sure this takes place for any incoming pushes.
1097 # However, do it here anyway just in case.
1098 make_packs_ugw objects/pack
1099
1100 # This is only effective with Git v2.3.5 and later and it will only matter when
1101 # we are using one of the "internal_rev_list" modes of pack-objects
1102 # (the combine-packs.sh script never uses any of those modes)
1103 # The "git repack" and "git prune" commands always sets this internally themselves
1104 # It makes no difference if there's no repository corruption
1105 GIT_REF_PARANOIA=1 && export GIT_REF_PARANOIA
1106
1107 # All of the options we might want to use with pack-objects were supported
1108 # at some point prior to Git version v1.6.6 which is the minimum version that
1109 # Girocco now requires.  Except for one (--use-bitmap-index).  Several of them
1110 # are "boiler plate" options we always want to use so we bundle them up here.
1111 pkopt="--delta-base-offset --keep-true-parents --non-empty --all-progress-implied"
1112 # We want to use --include-tag, but before Git v2.10.1 it would leave out
1113 # "middle" tags (e.g. a tag of a tag of a commit would omit the tagged tag)
1114 # See http://repo.or.cz/git.git/b773ddea2cd3b08c for details
1115 # ("pack-objects: walk tag chains for --include-tag", 2016-09-07, v2.10.1)
1116 # This is not a free check as it matches all refs against refs/tags/ then
1117 # peels all the annotated tags and checks for inclusion.  The situation in
1118 # which it would add a tag that was not already included by a reachability
1119 # trace that included tag starting points can only occur if a new tag gets
1120 # pushed during gc pointing to something that would have been packed anyway.
1121 # But, it could happen and, really, campared to gc as a whole it's not that
1122 # expensive to perform (provided we do not get an unconnected pack).
1123 [ -z "$var_have_git_2101" ] || pkopt="$pkopt --include-tag"
1124 pkopt="$pkopt ${quiet:---progress} $packopts"
1125
1126 # The git pack-objects command only supports bitmaps if all objects are being
1127 # packed (the "--all" option) and the "--stdout" option is NOT being used.
1128 # Additionally, while packing, if any encountered reachable objects are
1129 # determined to be "not wanted" then no bitmap index will be written anyway.
1130 # While it is theoretically possible that a project with a non-empty alternates
1131 # file ends up packing all objects (because it does not actually use any of the
1132 # objects found in the alternates), it's very unlikely.  And, in the unlikely
1133 # event that did occur, clients would see a message about only using one bitmap
1134 # because Git can only use one bitmap at a time and at least one of the
1135 # alternates is bound to have a bitmap.  Therefore if we see a non-empty
1136 # alternates file, we disable writing bitmaps which avoids the warning and any
1137 # possibility of a client warning as well.  Also if we are running anything
1138 # before Git v2.1.0 (the effective version for repack.writeBitmaps=true) then
1139 # we also always disable bitmap writing.
1140 wbmopt=
1141 [ -z "$var_have_git_210" ] || wbmopt="--write-bitmap-index"
1142 # More recent versions of pack-objects have optimizations when not using the
1143 # --local option.  If we do not have any alternates it's a pointless option.
1144 # If we do have alternates we need to skip writing a bitmap and we cannot
1145 # have a bundle since it must contain all objects.
1146 if [ -n "$isfork" ]; then
1147         lclopt="--local"
1148         wbmopt=
1149         makebndl=
1150 else
1151         lclopt=
1152         makebndl=1
1153 fi
1154
1155 #
1156 ## Phase I
1157 #
1158
1159 wbmstr=
1160 [ -n "$wbmopt" ] || wbmstr=" (bitmaps disabled)"
1161 progress "~ [$proj] running primary full gc pack-objects$wbmstr ($(date))"
1162
1163 gotforks=
1164 ! has_forks_with_alternates "$proj" || gotforks=1
1165
1166 # To avoid "Push Pack Redux" (see README-GC), after collecting the initial
1167 # preexisting non-keep pack list, we rename them so that an incoming push
1168 # pack cannot possibly experience a pack name collision.  Git does not require
1169 # use of the "default" pack names, simply that the proper extensions are used.
1170 # We rename to insert an "_r" just before the extension to avoid "Push Pack Redux"
1171 # name collisions.  Later on we may create an "unreachable" pack for hard-linking
1172 # down into forks and it will have an "_u" inserted just before its extension.
1173 packlist="$(list_packs -C objects/pack --all --exclude-no-idx --exclude-keep --quiet .)" || :
1174 oldpacks=
1175 for oldpack in $packlist; do
1176         oldpack="${oldpack%.pack}"
1177         [ -f "objects/pack/$oldpack.pack" ] || {
1178                 echo >&2 "[$proj] unable to list old pack files"
1179                 exit 1
1180         }
1181         case "$oldpack" in pre-auto-gc-[12])
1182                 # we never disturb pre-auto-gc-1 or pre-auto-gc-2 packs
1183                 continue
1184         esac
1185         oldpackhex="${oldpack#pack-}"
1186         if [ "${oldpackhex#*[!0-9a-fA-F]}" != "$oldpackhex" ]; then
1187                 # names not exclusively hexadecimal do not need renaming
1188                 oldpacks="${oldpacks:+$oldpacks }$oldpack"
1189                 continue
1190         fi
1191         rename_pack "objects/pack/$oldpack" "objects/pack/${oldpack}_r" || {
1192                 echo >&2 "[$proj] unable to rename old pack files"
1193                 exit 1
1194         }
1195         # If the oldpack has a .keep now it means a "Push Pack Redux" is actually
1196         # in progress at this moment and we need to .keep the renamed pack,
1197         # otherwise no "Push Pack Redux" has started yet or it has already finished.
1198         # In either case we're okay because if it's just finished then all ref
1199         # changes have already been made so we don't need a .keep and we will
1200         # see the ref changes and grab all the objects via a reachability trace.
1201         # If it hasn't started yet that's okay because we're done moving that
1202         # name so a complete pack will appear under the old name that we'll
1203         # leave alone.
1204         if [ -f "objects/pack/$oldpack.keep" ]; then
1205                 echo "Push Pack Redux" >"objects/pack/${oldpack}_r.keep"
1206         else
1207                 oldpacks="${oldpacks:+$oldpacks }${oldpack}_r"
1208         fi
1209 done
1210
1211 # We wish to keep deltas from our last full pack so if we're not redeltaing
1212 # then make sure the .pack associated with the .bitmap has a newer mod time
1213 # (If there is no .bitmap then touch the pack with the most objects instead.)
1214 if [ -z "$newdeltas" ]; then
1215         bmpack="$(list_packs --exclude-no-bitmap --exclude-no-idx --max-matches 1 objects/pack)"
1216         [ -n "$bmpack" ] || bmpack="$(list_packs --exclude-no-idx --max-matches 1 --object-limit -1 --include-boundary objects/pack)"
1217         if [ -n "$bmpack" ] && [ -f "$bmpack" ] && [ -s "$bmpack" ]; then
1218                 sleep 1
1219                 touch -c "$bmpack" 2>/dev/null || :
1220                 # We must touch .gc_in_progress here to avoid $bmpack looking
1221                 # like it's been "freshened" when redundant packs are removed
1222                 # It's okay if they have the same mod time, but POSIX does not
1223                 # guarantee an ordering for the "touching" that occurs which is
1224                 # why this must be a separate command but needs no "sleep 1"
1225                 touch .gc_in_progress
1226         fi
1227 fi
1228
1229 # Now we need to make sure that any "freshening" that takes place will actually
1230 # result in a "newer" modification time than the .gc_in_progress file now has
1231 sleep 1
1232
1233 # We run git pack-objects from the repack subdirectory so we can force
1234 # optimized packs to be generated even for repositories that do not have any
1235 # tagged commits
1236 packs="$(git --git-dir=repack pack-objects </dev/null \
1237         $pkopt --all $newdeltas $lclopt ${wbmopt:---honor-pack-keep} "$@" repack/alt/pack/pack)"
1238 vcnt packcnt $packs
1239 [ $packcnt -eq 1 ] || makebndl=
1240
1241 #
1242 ## Phase II
1243 #
1244
1245 progress "~ [$proj] running supplementary gc pack-objects ($(date))"
1246
1247 # Add the "supplementary" refs
1248 compute_extra_reachables >>repack/packed-refs
1249
1250 # Subtract the primary refs
1251 GIT_ALTERNATE_OBJECT_DIRECTORIES="$PWD/repack/alt"
1252 export GIT_ALTERNATE_OBJECT_DIRECTORIES
1253
1254 # For this one we MUST use --local and MUST NOT use --write-bitmap-index
1255 # However, if there is a "logs" subdirectory we need to use --reflog
1256 # We do add it, just in case, if the linked working trees dir is present
1257 # We do not add --indexed-objects as that requires v2.2.0 and it's unclear
1258 # if it properly includes linked working tree index files or not.  The
1259 # above compute_extra_reachables has already included all index trees (thereby
1260 # providing proper --indexed-objects support for all Git versions) making the
1261 # option completely unnecessary.
1262 rflopt=
1263 ! [ -d logs ] && ! [ -d worktrees ] || rflopt=--reflog
1264 spacks="$(git --git-dir=repack pack-objects </dev/null \
1265         $pkopt --honor-pack-keep --all $rflopt $newdeltas --local "$@" repack/alt/pack/pack)"
1266
1267 #
1268 ## Phase III
1269 #
1270
1271 # There's nothing to do for Phase III unless we have forks that refer to our
1272 # project from their alternates file
1273 hlpacks=
1274 upacks=
1275 if [ -n "$gotforks" ]; then
1276
1277         progress "~ [$proj] running keep-unreachable gc pack-objects for forks ($(date))"
1278
1279         # If we are a fork, any pre-existing _u packs need to have a .keep
1280         # for this phase and be added to the hlpacks list otherwise (we are
1281         # not a fork) pre-existing _u packs are anomalies to be treated like
1282         # regular non-_u packs
1283         if [ -n "$isfork" ]; then
1284                 for upack in $(find -L objects/pack -mindepth 1 -maxdepth 1 -name "pack-$octet20*_u.pack" -print); do
1285                         upack="${upack%.pack}"
1286                         [ -e "$upack.keep" ] || echo "unreachable" >"$upack.keep"
1287                         hlpacks="${hlpacks:+$hlpacks }${upack#objects/pack/pack-}"
1288                 done
1289         fi
1290         # Using either --no-reuse-delta or --no-reuse-object together with the
1291         # --keep-unreachable option is a very, very, very bad idea when good
1292         # packs are the desired outcome.  If newdeltas are being generated
1293         # then we pack to a temp name, and use combine-packs.sh to get a better
1294         # pack as the result to avoid making a bad --keep-unreachable pack
1295         pfx=
1296         [ -z "$newdeltas" ] || pfx="ku"
1297         upacks="$(git --git-dir=repack pack-objects </dev/null \
1298                 $pkopt --honor-pack-keep --all $rflopt --keep-unreachable --local "$@" repack/alt/pack/${pfx}pack)"
1299         if [ -n "$upacks" ] && [ -n "$newdeltas" ]; then
1300                 progress "~ [$proj] rebuilding keep-unreachable pack deltas"
1301                 oldupacks="$upacks"
1302                 upacks="$(
1303                         printf "repack/alt/pack/${pfx}pack-%s.pack\n" $oldupacks |
1304                         run_combine_packs --names --weak-naming --non-empty --all-progress-implied ${quiet:---progress} \
1305                                 $packopts $newdeltas "$@" repack/alt/pack/pack)"
1306                 eval rm -f "$(printf \""repack/alt/pack/${pfx}pack-%s.*"\"" " $oldupacks)"
1307         fi
1308         for upack in $upacks; do
1309                 rename_pack "repack/alt/pack/pack-$upack" "repack/alt/pack/pack-${upack}_u"
1310         done
1311         rm -f objects/pack/pack-*_u.keep
1312         [ -z "$hlpacks" ] && [ -z "$upacks" ] ||
1313         progress "~ [$proj] hard-linking keep-unreachable pack(s) into immediate child forks"
1314
1315         # We have to update the lastparentgc time in the child forks even if they do not get any
1316         # new "unreachable packs" because they need to run gc just in case the parent now has some
1317         # objects that used to only be in the child so they can be removed from the child.
1318         # For example, a "patch" might be developed first in a fork and then later accepted into
1319         # the parent in which case the objects making up the patch in the child fork are now
1320         # redundant (since they're now in the parent as well) and need to be removed from the
1321         # child fork which can only happen if the child fork runs gc.
1322         lastparentgc="$(date "$datefmt")"
1323
1324         # It is enough to copy objects just one level down and get_repo_list
1325         # takes a regular expression (which is automatically prefixed with '^')
1326         # so we can easily match forks exactly one level down from this project
1327         forkdir="$proj"
1328         get_repo_list "$forkdir/[^/:][^/:]*:" |
1329         while read fork; do
1330                 # Ignore forks that do not exist or are symbolic links
1331                 ! [ -L "$cfg_reporoot/$fork.git" ] && [ -d "$cfg_reporoot/$fork.git" ] ||
1332                         continue
1333                 # Or do not have a non-zero length alternates file
1334                 [ -s "$cfg_reporoot/$fork.git/objects/info/alternates" ] ||
1335                         continue
1336                 runupdate=
1337                 # Match hlpacks in parent project if any
1338                 if [ -n "$hlpacks" ]; then
1339                         mkdir -p "$cfg_reporoot/$fork.git/objects/pack"
1340                         eval ln -f "$(printf '"objects/pack/pack-%s.pack" ' $hlpacks)" \
1341                                 "$(printf '"objects/pack/pack-%s.idx" ' $hlpacks)" \
1342                                 '"$cfg_reporoot/$fork.git/objects/pack/"'
1343                         runupdate=1
1344                 fi
1345                 # Match upacks in repack/alt area if any
1346                 if [ -n "$upacks" ]; then
1347                         mkdir -p "$cfg_reporoot/$fork.git/objects/pack"
1348                         eval ln -f "$(printf '"repack/alt/pack/pack-%s_u.pack" ' $upacks)" \
1349                                 "$(printf '"repack/alt/pack/pack-%s_u.idx" ' $upacks)" \
1350                                 '"$cfg_reporoot/$fork.git/objects/pack/"'
1351                         runupdate=1
1352                 fi
1353                 if ! [ -e "$cfg_reporoot/$fork.git/.needsgc" ]; then
1354                         # Trigger a mini gc in the fork if it now has too many packs
1355                         packs="$(list_packs --quiet --count --exclude-no-idx --exclude-keep "$cfg_reporoot/$fork.git/objects/pack")" || :
1356                         if [ -n "$packs" ] && [ "$packs" -ge 20 ]; then
1357                                 >"$cfg_reporoot/$fork.git/.needsgc"
1358                         fi
1359                 fi
1360                 [ -z "$runupdate" ] || git --git-dir="$cfg_reporoot/$fork.git" update-server-info
1361                 # Update the fork's lastparentgc date (must be more recent than $gcstart)
1362                 git --git-dir="$cfg_reporoot/$fork.git" config gitweb.lastparentgc "$lastparentgc"
1363         done
1364 fi
1365
1366 # Now move any primary/supplementary packs back into objects/pack
1367 # then drop any "unfreshened" redundant packs and clear repack/alt
1368
1369 # First make sure the primary pack(s) have the most recent mod time
1370 if [ -n "$packs" ]; then
1371         [ -z "$spacks" ] || sleep 1
1372         printf 'repack/alt/pack/pack-%s.pack\n' $packs | xargs touch -c 2>/dev/null || :
1373 fi
1374
1375 # Move the packs into place
1376 for pack in $packs $spacks; do
1377         rename_pack "repack/alt/pack/pack-$pack" "objects/pack/pack-$pack"
1378 done
1379
1380 # It's possible that one of the $oldpacks had a .bitmap, got renamed (along
1381 # with its .bitmap) and then got "freshened" causing us to not remove it
1382 # However, if $wbmopt is set we most likely now have TWO .bitmap packs!
1383 # This can produce ugly warnings we don't want and possibly get the wrong
1384 # bitmap used since only one .bitmap file can ever be used by Git.
1385 # If this has happened, the .bitmap we want to discard will always have
1386 # an _r infix so we can just zap any such now since it will leave the pack.
1387 [ -z "$wbmopt" ] || rm -f objects/pack/pack-*_r.bitmap || :
1388
1389 # Remove the redundant packs that have not since been "freshened"
1390 # This does not completely eliminate the race condition window (Girocco's own
1391 # activites -- gc/fetch/receive are immune to the race) but it substantially
1392 # shrinks it down to just the time after the find but before the following rm
1393 >repack/oldpacks
1394 [ -z "$oldpacks" ] ||
1395 printf 'objects/pack/%s.pack\n' $oldpacks |
1396 LC_ALL=C sort >repack/oldpacks
1397 find -L objects/pack -maxdepth 1 -type f -name "pack-$octet20*.pack" -newer .gc_in_progress -print |
1398 LC_ALL=C sort >repack/freshened
1399 deadpacks="$(LC_ALL=C join -v 1 repack/oldpacks repack/freshened | LC_ALL=C sed 's/\.pack$//')"
1400 [ -z "$deadpacks" ] ||
1401 eval echo "$(printf '"%s".* ' $deadpacks)" | xargs rm -f || :
1402
1403 # No need for this anymore
1404 rm -rf repack/alt objects/pack/repack
1405 unset GIT_ALTERNATE_OBJECT_DIRECTORIES
1406
1407 #
1408 ## Phase IV
1409 #
1410
1411 progress "~ [$proj] running gc prune-packed"
1412
1413 # We do not want the redundant packs or any new "--keep-unreachable" pack(s) to be
1414 # present while running prune-packed.  We try to guarantee that any loose object
1415 # that's unreachable persists for at least one $Girocco::Config::min_gc_interval
1416 # (not withstanding administrator interference to force earlier gc to occur).
1417 # If we were to include the redundant/keep-unreachable pack(s) when running
1418 # prune-packed and a loose unreachable object happened to be duplicated in one
1419 # of them we would end up removing it too soon and void our guarantee.
1420 git prune-packed $quiet
1421
1422 progress "~ [$proj] running loose objects gc pack-objects ($(date))"
1423
1424 # Although Git v2.10.0 and later support a --pack-loose-unreachable option,
1425 # we MUST NOT use it for these reasons:
1426 #  1) We're not interested in expensive "unreachable" at this point, only "loose"
1427 #  2) It produces simply horrid packs about 3.8x times larger than they should be
1428 #  3) We don't require anything more than Git v1.6.6
1429 lpacks="$(run_combine_packs </dev/null --names --loose --weak-naming --non-empty --all-progress-implied ${quiet:---progress} $packopts $newdeltas "$@")"
1430
1431 if [ -n "$lpacks" ]; then
1432         # Make sure any primary pack(s) have a more recent mod time than "unreachable" objects packs
1433         if [ -n "$packs" ]; then
1434                 sleep 1
1435                 printf 'objects/pack/pack-%s.pack\n' $packs | xargs touch -c 2>/dev/null || :
1436         fi
1437         # We need to identify these packs later so we don't combine_packs them
1438         for objpack in $lpacks; do
1439                 rename_pack "objects/pack/pack-$objpack" "objects/pack/pack-${objpack}_o" || :
1440         done
1441         # Finally zap the corresponding loose objects
1442         progress "~ [$proj] running packed loose objects gc prune-packed"
1443         git prune-packed $quiet
1444 fi
1445
1446 ! [ -e .gc_failed ] || exit 1
1447 # These, if they exist, are now meaningless and need to be removed
1448 rm -f gfi-packs .needsgc .svnpack .svnpackgc
1449
1450 # Make sure this stays up to date
1451 git update-server-info
1452
1453 # We must make loose objects group writable so that they
1454 # can be freshened by other pushers.  Technically we need only do this for
1455 # push projects but to enable mirror projects to be more easily converted to
1456 # push projects, we go ahead and do it for all projects.
1457 # By the time we get here we really shouldn't have any of these, but just in case.
1458 { find -L objects/$octet -type f -name "$octet19*" -exec chmod ug+w '{}' + || :; } 2>/dev/null
1459
1460 # darcs:// mirrors have a xxx.log file that will grow endlessly
1461 # if this is a mirror and the file exists, shorten it to 10000 lines
1462 # also take this opportunity to optimize the darcs repo
1463 if ! [ -e .nofetch ] && [ -n "$cfg_mirror" ]; then
1464         url="$(config_get baseurl)" || :
1465         case "$url" in darcs://*)
1466                 if [ -n "$cfg_mirror_darcs" ]; then
1467                         url="${url%/}"
1468                         basedarcs="$(basename "${url#darcs:/}")"
1469                         if [ -f "$basedarcs.log" ]; then
1470                                 tail -n 10000 "$basedarcs.log" >"$basedarcs.log.$$"
1471                                 mv -f "$basedarcs.log.$$" "$basedarcs.log"
1472                         fi
1473                         if [ -d "$basedarcs.darcs" ]; then
1474                                 (
1475                                         cd "$basedarcs.darcs"
1476                                         # without show_progress suppress non-error output
1477                                         [ -n "$show_progress" ] || exec >/dev/null
1478                                         # Note that this does not optimize _darcs/inventories/ :(
1479                                         darcs optimize || :
1480                                 )
1481                         fi
1482                 fi
1483         esac
1484 fi
1485
1486 # Create a matching .bndl header file for the all-in-one pack we just created
1487 # but only if we're not a fork (otherwise the bundle would not be complete)
1488 # and we are running at least Git version 1.7.2 (pack_is_complete always fails otherwise)
1489 if [ -n "$makebndl" ] && [ -n "$var_have_git_172" ]; then
1490         # There should only be one pack in $packs but do some checking...
1491         # The one we just created will have a .idx and will NOT have a .keep
1492         progress "~ [$proj] creating downloadble bundle header"
1493         pkbase=
1494         pkhead=
1495         IFS= read -r curhead <repack/HEAD.orig || :
1496         if
1497                 [ -s "objects/pack/pack-$packs.pack" ] &&
1498                 [ -s "objects/pack/pack-$packs.idx" ] &&
1499                 ! [ -e "objects/pack/pack-$packs.keep" ] &&
1500                 pkhead="$(pack_is_complete "$PWD/objects/pack/pack-$packs.pack" \
1501                         "$PWD/repack/packed-refs.orig" "$curhead")"
1502         then
1503                 pkbase="objects/pack/pack-$packs"
1504         fi
1505         if [ -n "$pkbase" ] && [ -n "$pkhead" ]; then
1506                 {
1507                         symref=
1508                         case "$curhead" in "ref: refs/"?*|"ref:refs/"?*|"refs/"?*)
1509                                 symref="${curhead#ref:}"
1510                                 symref="${symref# }"
1511                         esac
1512                         bndlurl=
1513                         [ -z "$cfg_httpbundleurl" ] || bndlurl=" url=$cfg_httpbundleurl/$proj.git/clone.bundle"
1514                         echo "# v2 git bundle"
1515                         LC_ALL=C sed -ne "/^$octet20$hexdig* refs\/[^ $tab]*\$/ p" <repack/packed-refs.orig
1516                         if [ -n "$symref" ]; then
1517                                 printf "$pkhead HEAD\0symref=HEAD:%s%s\n" "$symref" "$bndlurl"
1518                         else
1519                                 if [ -n "$bndlurl" ]; then
1520                                         printf "$pkhead HEAD\0%s\n" "${bndlurl:# }"
1521                                 else
1522                                         echo "$pkhead HEAD"
1523                                 fi
1524                         fi
1525                         echo ""
1526                 } >"$pkbase.bndl"
1527                 bndletag="$("$cfg_basedir/bin/rangecgi" --etag -m 1 "$pkbase.bndl" "$pkbase.pack")" || :
1528                 bndlsha="$(printf '%s' "$bndletag" | git hash-object --stdin)" || :
1529                 if [ -n "$bndletag" ]; then
1530                         case "$bndlsha" in $octet20*)
1531                                 bndlshatrailer="${bndlsha#????????}"
1532                                 bndlshaprefix="${bndlsha%$bndlshatrailer}"
1533                                 bndlname="$(TZ=UTC date +%Y%m%d_%H%M%S)-${bndlshaprefix:-0}"
1534                                 [ -d bundles ] || mkdir bundles
1535                                 echo "${pkbase#objects/pack/}.bndl" >"bundles/$bndlname"
1536                                 echo "${pkbase#objects/pack/}.pack" >>"bundles/$bndlname"
1537                                 ln -s -f -n "$bndlname" bundles/latest
1538                         esac
1539                 fi
1540         fi
1541 fi
1542
1543 # Record the size of this repo as the sum of its clone packed-refs + *.pack sizes as 1024-byte blocks
1544 eval "reposizek=$(( $(
1545         echo 0 $(du -k repack/packed-refs.orig $(printf 'objects/pack/pack-%s.pack ' $packs) 2>/dev/null |
1546                 LC_ALL=C awk '{print $1}') |
1547         LC_ALL=C sed -e 's/ / + /g') ))"
1548 config_set_raw girocco.reposizek "${reposizek:-0}"
1549
1550 # Now we're finally done with this
1551 rm -rf repack
1552
1553 # We didn't used to do anything about rerere or worktrees but we're
1554 # trying to make nice with linked working trees these days :)
1555 # Maybe even non-bare repositories too, but *shush* about those ;)
1556 if [ -n "$var_have_git_250" ] && [ -d worktrees ]; then
1557         # The value "3.months.ago" is hard-coded into gc.c rather than
1558         # having the default be in worktree.c so we must provide it if
1559         # we get nothing out of the gc.worktreePruneExpire config item
1560         # Prior to Git v2.6.0 the config item was gc.pruneworktreesexpire
1561         # however we just always use the newer name no matter what Git version
1562         expiry="$(git config --get gc.worktreePruneExpire 2>/dev/null)" || :
1563         eval git worktree prune --expire '"${expiry:-3.months.ago}"' "${quiet:+>/dev/null 2>&1}" || :
1564 fi
1565 # git rerere does it right and handles its own default/config'd expiration values
1566 ! [ -d rr-cache ] || eval git rerere gc "${quiet:+>/dev/null 2>&1}" || :
1567
1568 # We use $gcstart here to avoid a race where a push occurs during the gc itself
1569 # and the next future gc could be incorrectly skipped if we used the current
1570 # timestamp here instead
1571 config_set lastgc "$gcstart"
1572 rm -f "$lockf"
1573
1574 progress "- [$proj] garbage check ($(date))"