I have a vacuum process that is sitting around and apparently not doing
anything. It's been around over 2000 seconds and is eating up no cpu.
It isn't waiting on a lock. Backtrace is this:
#0 0x000000367aed4ff7 in semop () from /lib64/libc.so.6
#1 0x00000000005d2a83 in PGSemaphoreLock (sema=0x2ace8a5721f0,
interruptOK=1 '\001') at pg_sema.c:418
#2 0x00000000005fa275 in LockBufferForCleanup (buffer=138743) at
bufmgr.c:2471
#3 0x0000000000476ff0 in btvacuumscan (info=0x7fff3b3dcfc0,
stats=0x1eb0a900, callback=0x551250 <lazy_tid_reaped>,
callback_state=0x1eae2db0, cycleid=45547) at nbtree.c:856
#4 0x00000000004774c5 in btbulkdelete (fcinfo=<value optimized out>) at
nbtree.c:574
#5 0x00000000006d100a in FunctionCall4Coll (flinfo=0xfffffffffffffffc,
collation=993900848, arg1=1, arg2=18446744073709551615, arg3=0,
arg4=140734187288784) at fmgr.c:1371
#6 0x00000000005511c4 in lazy_vacuum_index (indrel=0x2ace8b697ab8,
stats=0x1eb09cc8, vacrelstats=0x1eae2db0) at vacuumlazy.c:1020
#7 0x0000000000551a77 in lazy_vacuum_rel (onerel=0x2ace8b696470,
vacstmt=<value optimized out>, bstrategy=<value optimized out>) at
vacuumlazy.c:843
#8 0x00000000005507e0 in vacuum_rel (relid=3471206537,
vacstmt=0x7fff3b3de740, do_toast=<value optimized out>, for_wraparound=0
'\0') at vacuum.c:1042
#9 0x0000000000550b0b in vacuum (vacstmt=0x7fff3b3de740, relid=<value
optimized out>, do_toast=0 '\0', bstrategy=<value optimized out>,
for_wraparound=0 '\0', isTopLevel=<value optimized out>) at vacuum.c:227
#10 0x00000000005d5d57 in do_autovacuum () at autovacuum.c:2739
#11 0x00000000005d6393 in AutoVacWorkerMain (argc=<value optimized out>,
argv=<value optimized out>) at autovacuum.c:1641
#12 0x00000000005d6486 in StartAutoVacWorker () at autovacuum.c:1428
#13 0x00000000005e0d22 in sigusr1_handler (postgres_signal_arg=<value
optimized out>) at postmaster.c:4588
#14 <signal handler called>
#15 0x000000367aecce83 in __select_nocancel () from /lib64/libc.so.6
#16 0x00000000005dd7dc in ServerLoop () at postmaster.c:1415
#17 0x00000000005dec97 in PostmasterMain (argc=2, argv=0x1ea0bd90) at
postmaster.c:1116
#18 0x000000000058352e in main (argc=2, argv=<value optimized out>) at
main.c:199
So what is this doing? This is a problem because other threads are
blocking on this one, and as soon as I kill -INT it it respawns so I
can't get rid of it.
pg version is:
PostgreSQL 9.1.2 on x86_64-unknown-linux-gnu, compiled by gcc (GCC)
4.1.2 20080704 (Red Hat 4.1.2-46), 64-bit
