Re: Mysteriously varying index scan costs

On Mon, Sep 24, 2018 at 2:20 PM, Laurenz Albe wrote: > > > But how can it be that the first run has to touch 74917 blocks, > while whe second run only needs to touch 1185? > > The first index scan may have killed lots of index tuples. Thanks, Pavan -- Pavan Deolasee

Re: pg_locks.relation question

running in Session 1 and then query pg_class for those OIDs, you might see more details. Of course, I am just guessing without looking into much detail. Thanks, Pavan -- Pavan Deolasee http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Training & Services

Re: ERROR: found multixact from before relminmxid

y map bit is set, thus leaving behind an unfrozen multixid. The page then again becomes !all_visible and the next vacuum then complains about the unfrozen multixid. Thanks, Pavan -- Pavan Deolasee http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Training & Services

Re: ERROR: unexpected chunk number 0 (expected 1) for toast value 76753264 in pg_toast_10920100

On Fri, Apr 6, 2018 at 8:55 AM, Pavan Deolasee wrote: > > > On Fri, Apr 6, 2018 at 2:34 AM, Tom Lane wrote: > >> a...@novozymes.com (Adam =?utf-8?Q?Sj=C3=B8gren?=) writes: >> >> [... still waiting for the result, I will return with what it said >> >>

Re: ERROR: unexpected chunk number 0 (expected 1) for toast value 76753264 in pg_toast_10920100

a reproducible test case to demonstrate the problem and writing the fix. More details on that soon. Thanks, Pavan -- Pavan Deolasee http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Training & Services