Re: shmem_seq may be a bad idea - Mailing list pgsql-hackers

From Thomas Lockhart
Subject Re: shmem_seq may be a bad idea
Date
Msg-id 390EB439.E42750F6@alumni.caltech.edu
Whole thread Raw
In response to Re: shmem_seq may be a bad idea  (Peter Eisentraut <e99re41@DoCS.UU.SE>)
Responses Re: shmem_seq may be a bad idea  (Tom Lane <tgl@sss.pgh.pa.us>)
Re: shmem_seq may be a bad idea  (Peter Eisentraut <peter_e@gmx.net>)
List pgsql-hackers
> A while ago while thinking about a way to make ipcclean better I thunk
> that perhaps the postmaster should write the keys of the segments it gets
> to a flat-text file. If it somehow crashes and loses track of what it
> allocated before it can use that information to clean up. Not sure how
> often that would take effect but it's very socially friendly.

Hmm. Could we write this to a separate shared memory segment? Much
more likely to be of fixed length and compatible between versions, and
more likely to exist or not exist with the same behavior as the large
shared memory segment under discussion??
                       - Thomas

-- 
Thomas Lockhart                lockhart@alumni.caltech.edu
South Pasadena, California


pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: Request for 7.0 JDBC status
Next
From: Bruce Momjian
Date:
Subject: Re: PgAccess 0.98.6 , a ugly bug fixed