| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Vadim Mikheev <vadim(at)krs(dot)ru> |
| Cc: | Hiroshi Inoue <Inoue(at)tpf(dot)co(dot)jp>, t-ishii(at)sra(dot)co(dot)jp, PostgreSQL-development <pgsql-hackers(at)postgreSQL(dot)org> |
| Subject: | Re: [HACKERS] Open 6.5 items |
| Date: | 1999-05-31 15:03:18 |
| Message-ID: | 8928.928162998@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Vadim Mikheev <vadim(at)krs(dot)ru> writes:
>> As for my test case,xidHash is filled with XactLockTable entries which have
>> been acquired by XactLockTableWait().
>> Could those entries be released immediately after they are acquired ?
> Ops. Thanks! Must be released.
Does this account for the "ShmemAlloc: out of memory" errors we've been
seeing? I spent a good deal of time yesterday grovelling through all
the calls to ShmemAlloc, and concluded that (unless there is a memory
stomp somewhere) it has to be caused by one of the shared hashtables
growing well beyond its size estimate.
I did find that the PROC structures are not counted in the initial
sizing of the shared memory block. This is no problem at the default
limit of 32 backends, but could get to be an issue for hundreds of
backends. I will add that item to the size estimate today.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 1999-05-31 15:12:07 | Re: [HACKERS] Backend sent 0x45 type while idle |
| Previous Message | Tom Lane | 1999-05-31 14:56:48 | Re: [HACKERS] Backends waiting, spinlocks, shared mem patches |