Re: Subscription statistics are not dropped at DROP SUBSCRIPTION in some cases

2023-09-08 Thread Masahiko Sawada
On Thu, Sep 7, 2023 at 10:22 PM Masahiko Sawada wrote: > > On Tue, Sep 5, 2023 at 11:32 AM Zhijie Hou (Fujitsu) > wrote: > > > > On Monday, September 4, 2023 10:42 PM Masahiko Sawada > > wrote: > > > > Hi, > > > > > On Mon, Sep 4, 2023 at 9:38 PM Zhijie Hou (Fujitsu) > > > > > > wrote: > > >

Re: Subscription statistics are not dropped at DROP SUBSCRIPTION in some cases

2023-09-07 Thread Masahiko Sawada
On Tue, Sep 5, 2023 at 11:32 AM Zhijie Hou (Fujitsu) wrote: > > On Monday, September 4, 2023 10:42 PM Masahiko Sawada > wrote: > > Hi, > > > On Mon, Sep 4, 2023 at 9:38 PM Zhijie Hou (Fujitsu) > > wrote: > > > > > > On Wednesday, July 5, 2023 2:53 PM Masahiko Sawada > > wrote: > > > > > > > Th

RE: Subscription statistics are not dropped at DROP SUBSCRIPTION in some cases

2023-09-04 Thread Zhijie Hou (Fujitsu)
On Monday, September 4, 2023 10:42 PM Masahiko Sawada wrote: Hi, > On Mon, Sep 4, 2023 at 9:38 PM Zhijie Hou (Fujitsu) > wrote: > > > > On Wednesday, July 5, 2023 2:53 PM Masahiko Sawada > wrote: > > > > > Thanks for reviewing the patch. Pushed. > > > > My colleague Vignesh noticed that the n

Re: Subscription statistics are not dropped at DROP SUBSCRIPTION in some cases

2023-09-04 Thread Masahiko Sawada
Hi, On Mon, Sep 4, 2023 at 9:38 PM Zhijie Hou (Fujitsu) wrote: > > On Wednesday, July 5, 2023 2:53 PM Masahiko Sawada > wrote: > > Hi, > > > Thanks for reviewing the patch. Pushed. > > My colleague Vignesh noticed that the newly added test cases were failing in > BF animal sungazer[1]. Thank

RE: Subscription statistics are not dropped at DROP SUBSCRIPTION in some cases

2023-09-04 Thread Zhijie Hou (Fujitsu)
On Wednesday, July 5, 2023 2:53 PM Masahiko Sawada wrote: Hi, > Thanks for reviewing the patch. Pushed. My colleague Vignesh noticed that the newly added test cases were failing in BF animal sungazer[1]. The test failed to drop the slot which is active on publisher. --error-log-- This failu

Re: Subscription statistics are not dropped at DROP SUBSCRIPTION in some cases

2023-07-04 Thread Masahiko Sawada
On Mon, Jun 19, 2023 at 12:37 PM Amit Kapila wrote: > > On Mon, Jun 19, 2023 at 6:50 AM Masahiko Sawada wrote: > > > > On Sat, Jun 17, 2023 at 6:45 PM Amit Kapila wrote: > > > > > > On Tue, May 16, 2023 at 8:00 PM Masahiko Sawada > > > wrote: > > > > > > > > On Thu, May 11, 2023 at 5:12 PM Mas

Re: Subscription statistics are not dropped at DROP SUBSCRIPTION in some cases

2023-06-18 Thread Amit Kapila
On Mon, Jun 19, 2023 at 6:50 AM Masahiko Sawada wrote: > > On Sat, Jun 17, 2023 at 6:45 PM Amit Kapila wrote: > > > > On Tue, May 16, 2023 at 8:00 PM Masahiko Sawada > > wrote: > > > > > > On Thu, May 11, 2023 at 5:12 PM Masahiko Sawada > > > wrote: > > > > > > > > > > After thinking more abo

Re: Subscription statistics are not dropped at DROP SUBSCRIPTION in some cases

2023-06-18 Thread Masahiko Sawada
On Sat, Jun 17, 2023 at 6:45 PM Amit Kapila wrote: > > On Tue, May 16, 2023 at 8:00 PM Masahiko Sawada wrote: > > > > On Thu, May 11, 2023 at 5:12 PM Masahiko Sawada > > wrote: > > > > > > > After thinking more about it, I realized that this is not a problem > > specific to HEAD. ISTM the probl

Re: Subscription statistics are not dropped at DROP SUBSCRIPTION in some cases

2023-06-17 Thread Amit Kapila
On Tue, May 16, 2023 at 8:00 PM Masahiko Sawada wrote: > > On Thu, May 11, 2023 at 5:12 PM Masahiko Sawada wrote: > > > > After thinking more about it, I realized that this is not a problem > specific to HEAD. ISTM the problem is that by commit 7b64e4b3, we drop > the stats entry of subscription

Re: Subscription statistics are not dropped at DROP SUBSCRIPTION in some cases

2023-05-16 Thread Masahiko Sawada
On Thu, May 11, 2023 at 5:12 PM Masahiko Sawada wrote: > > On Wed, May 10, 2023 at 8:58 PM Melih Mutlu wrote: > > > > Hi, > > > > Masahiko Sawada , 8 May 2023 Pzt, 10:24 tarihinde > > şunu yazdı: > >> > >> I've attached the patch. Feedback is very welcome. > > > > > > Thanks for the patch, nice

Re: Subscription statistics are not dropped at DROP SUBSCRIPTION in some cases

2023-05-11 Thread Masahiko Sawada
On Wed, May 10, 2023 at 8:58 PM Melih Mutlu wrote: > > Hi, > > Masahiko Sawada , 8 May 2023 Pzt, 10:24 tarihinde şunu > yazdı: >> >> I've attached the patch. Feedback is very welcome. > > > Thanks for the patch, nice catch. > I can confirm that the issue exists on HEAD and gets resolved by this p

Re: Subscription statistics are not dropped at DROP SUBSCRIPTION in some cases

2023-05-10 Thread Melih Mutlu
Hi, Masahiko Sawada , 8 May 2023 Pzt, 10:24 tarihinde şunu yazdı: > I've attached the patch. Feedback is very welcome. > Thanks for the patch, nice catch. I can confirm that the issue exists on HEAD and gets resolved by this patch. Also it looks like stats are really not affected if transaction

RE: Subscription statistics are not dropped at DROP SUBSCRIPTION in some cases

2023-05-08 Thread Hayato Kuroda (Fujitsu)
Dear Sawada-san, Thank you for giving the patch! I confirmed that the problem you raised could be occurred on the HEAD, and the test you added could reproduce that. When the stats entry has been removed but pg_stat_get_subscription_stats() is called, the returned values are set as 0x0. Additiona

Re: Subscription statistics are not dropped at DROP SUBSCRIPTION in some cases

2023-05-08 Thread Nathan Bossart
On Mon, May 08, 2023 at 04:23:15PM +0900, Masahiko Sawada wrote: > We call pgstat_drop_subscription() at the end of DropSubscription() > but we could leave from this function earlier e.g. when no slot is > associated with the subscription. In this case, the statistics entry > for the subscription r

Subscription statistics are not dropped at DROP SUBSCRIPTION in some cases

2023-05-08 Thread Masahiko Sawada
Hi, We call pgstat_drop_subscription() at the end of DropSubscription() but we could leave from this function earlier e.g. when no slot is associated with the subscription. In this case, the statistics entry for the subscription remains. To fix it, I think we need to call it earlier, just after re