Hi Tom, Alvaro,Thanks for your work on this. Please keep me posted as to which version in CVS this fix will be applied to and I will do my best to test it.Thanks again! --Dylan HansenEnterprise Systems DeveloperOn 27-Jun-06, at 5:42 AM, Alvaro Herrera wrote:Tom Lane wrote: Alvaro Herrera <[EMAIL P
, at 4:09 PM, Matthew T. O'Connor wrote:Tom Lane wrote: Dylan Hansen <[EMAIL PROTECTED]> writes: I have been spending some time looking into how auto-vacuum is performing on one of our servers. After putting the PostgreSQL logs in debug I noticed that the threshold for ANALYZE was
Hi Tom, thanks for your reply.On 23-Jun-06, at 11:17 AM, Tom Lane wrote:--oids is only meant to preserve OIDs within user tables; it never hasand never will preserve OIDs for system-catalog entries.So just to clarify, the table's OID itself will never be preserved, but the data inside the table wil
Greetings everyone,Since we've started using the pg_autovacuum table we've come to realize that keeping OID values between our database dumps is critical. I've been doing some testing using the pg_dump command with the --oids option. For some reason, it doesn't seem like the the OID values are be
Hi Tom, Alvaro. Thanks so much for your replies.On 22-Jun-06, at 12:56 PM, Alvaro Herrera wrote:Hmm ... I'd think that the number of dead tuples plus live tuples shouldnever be smaller than the number of tuples seen at last analyze. Unlesssome stats messages are lost (say, stop the server, start
Greetings all,I have been spending some time looking into how auto-vacuum is performing on one of our servers. After putting the PostgreSQL logs in debug I noticed that the threshold for ANALYZE was never being hit for a particular table because the calculated value becomes increasingly negative.W
