Re: Improve checking for pg_index.xmin

2020-03-24 Thread Alexander Korotkov
On Tue, Mar 24, 2020 at 3:38 PM Amit Kapila wrote: > On Sun, Jan 12, 2020 at 3:33 AM Alexander Korotkov > wrote: > > > > On Wed, Jan 8, 2020 at 4:37 PM Tom Lane wrote: > > > Heikki Linnakangas writes: > > > > On 01/11/2019 01:50, Alexander Korotkov wrote: > > > >> This happens so, because we're

Re: Improve checking for pg_index.xmin

2020-03-24 Thread Amit Kapila
On Sun, Jan 12, 2020 at 3:33 AM Alexander Korotkov wrote: > > On Wed, Jan 8, 2020 at 4:37 PM Tom Lane wrote: > > Heikki Linnakangas writes: > > > On 01/11/2019 01:50, Alexander Korotkov wrote: > > >> This happens so, because we're checking that there is no broken HOT > > >> chains after index cr

Re: Improve checking for pg_index.xmin

2020-01-11 Thread Alexander Korotkov
On Wed, Jan 8, 2020 at 4:37 PM Tom Lane wrote: > Heikki Linnakangas writes: > > On 01/11/2019 01:50, Alexander Korotkov wrote: > >> This happens so, because we're checking that there is no broken HOT > >> chains after index creation by comparison pg_index.xmin and > >> TransactionXmin. So, we c

Re: Improve checking for pg_index.xmin

2020-01-08 Thread Tom Lane
Heikki Linnakangas writes: > On 01/11/2019 01:50, Alexander Korotkov wrote: >> This happens so, because we're checking that there is no broken HOT >> chains after index creation by comparison pg_index.xmin and >> TransactionXmin. So, we check that pg_index.xmin is in the past for >> current tran

Re: Improve checking for pg_index.xmin

2020-01-08 Thread Heikki Linnakangas
On 01/11/2019 01:50, Alexander Korotkov wrote: Hi! Our customer faced with issue, when index is invisible after creation. The reproducible case is following. $ psql db2 # begin; # select txid_current(); $ psql db1 # select i as id, 0 as v into t from generate_series(1, 10) i;