Strange notices, should I worry?

Ashley Clark <aclark@ghoti.org> writes:

Occasionally and without warning I get this from my daily vacuum
cronjob:
NOTICE: RegisterSharedInvalid: SI buffer overflow
NOTICE: InvalidateSharedInvalid: cache state reset
I don't understand what these mean. Should I be concerned about them
and what do they signify?

No real need to worry. Those should've been downgraded to DEBUG-level
messages a release or two back, but nobody bothered...

regards, tom lane

Ashley Clark <aclark@ghoti.org> writes:

Occasionally and without warning I get this from my daily vacuum
cronjob:
NOTICE: RegisterSharedInvalid: SI buffer overflow
NOTICE: InvalidateSharedInvalid: cache state reset
I don't understand what these mean. Should I be concerned about them
and what do they signify?

No real need to worry. Those should've been downgraded to DEBUG-level
messages a release or two back, but nobody bothered...

OK, Tom, I assume you are recommending the following changes. I will
apply the patch.

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026

* Bruce Momjian in "Re: [GENERAL] Strange notices, should I worry?"
* dated 2001/05/07 13:19 wrote:

No real need to worry. Those should've been downgraded to
DEBUG-level messages a release or two back, but nobody bothered...

OK, Tom, I assume you are recommending the following changes. I will
apply the patch.

If it's not too much of a bother, what do they mean?

--
shaky caller

Checking application/pgp-signature: FAILURE
-- Start of PGP signed section.

* Bruce Momjian in "Re: [GENERAL] Strange notices, should I worry?"
* dated 2001/05/07 13:19 wrote:

No real need to worry. Those should've been downgraded to
DEBUG-level messages a release or two back, but nobody bothered...

OK, Tom, I assume you are recommending the following changes. I will
apply the patch.

If it's not too much of a bother, what do they mean?

OK, when one backend changes a system table, it invalidates the system
caches used by other backends. I don't remember the exact messages, but
they deals with cache invalidation overflows that are properly handled
by each backend and do not present any cause for concern.

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026

Bruce Momjian <pgman@candle.pha.pa.us> writes:

If it's not too much of a bother, what do they mean?

OK, when one backend changes a system table, it invalidates the system
caches used by other backends. I don't remember the exact messages, but
they deals with cache invalidation overflows that are properly handled
by each backend and do not present any cause for concern.

What they really mean is that some backend has failed to keep up with
the stream of cache invalidation messages being issued by some other
backend. The system recovers from this by simply invalidating *all*
cache entries when it happens. The reason there's a message is

(1) a version or three back, there were various bugs that might be
triggered by these events. AFAIK they're all fixed.

(2) if you see this a *lot* (not just occasionally) it might suggest
that the cache inval message buffer size ought to be increased. So
someone once felt there was value in notifying the user.

regards, tom lane