Re: Deadlock between backend and recovery may not be detected

2021-01-05 Thread Fujii Masao
On 2021/01/06 11:48, Masahiko Sawada wrote: On Tue, Jan 5, 2021 at 3:26 PM Fujii Masao wrote: On 2020/12/25 13:16, Kyotaro Horiguchi wrote: At Wed, 23 Dec 2020 21:42:47 +0900, Fujii Masao wrote in you. Attached is the updated of the patch. What about this version? The patch contains

Re: Deadlock between backend and recovery may not be detected

2021-01-05 Thread Masahiko Sawada
On Tue, Jan 5, 2021 at 3:26 PM Fujii Masao wrote: > > > > On 2020/12/25 13:16, Kyotaro Horiguchi wrote: > > At Wed, 23 Dec 2020 21:42:47 +0900, Fujii Masao > > wrote in > >> you. Attached > >> is the updated of the patch. What about this version? > > > > The patch contains a hunk in the followin

Re: Deadlock between backend and recovery may not be detected

2021-01-05 Thread Kyotaro Horiguchi
At Tue, 5 Jan 2021 15:26:50 +0900, Fujii Masao wrote in > > > On 2020/12/25 13:16, Kyotaro Horiguchi wrote: > > At Wed, 23 Dec 2020 21:42:47 +0900, Fujii Masao > > wrote in > >> you. Attached > >> is the updated of the patch. What about this version? > > The patch contains a hunk in the follo

Re: Deadlock between backend and recovery may not be detected

2021-01-05 Thread Victor Yegorov
вт, 5 янв. 2021 г. в 07:26, Fujii Masao : > This situation makes me feel that I'm inclined to skip the back-patch to > v9.5. > Because the next minor version release is the final one for v9.5. So if we > unexpectedly introduce the bug to v9.5 by the back-patch, there is no > chance to fix that. OT

Re: Deadlock between backend and recovery may not be detected

2021-01-04 Thread Fujii Masao
On 2020/12/25 13:16, Kyotaro Horiguchi wrote: At Wed, 23 Dec 2020 21:42:47 +0900, Fujii Masao wrote in you. Attached is the updated of the patch. What about this version? The patch contains a hunk in the following structure. + if (got_standby_lock_timeout) + goto clea

Re: Deadlock between backend and recovery may not be detected

2020-12-24 Thread Kyotaro Horiguchi
At Wed, 23 Dec 2020 21:42:47 +0900, Fujii Masao wrote in > you. Attached > is the updated of the patch. What about this version? The patch contains a hunk in the following structure. + if (got_standby_lock_timeout) + goto cleanup; + + if (got_standby_deadlock_timeout)

Re: Deadlock between backend and recovery may not be detected

2020-12-24 Thread Masahiko Sawada
On Wed, Dec 23, 2020 at 9:42 PM Fujii Masao wrote: > > > > On 2020/12/23 19:28, Masahiko Sawada wrote: > > On Tue, Dec 22, 2020 at 11:58 PM Fujii Masao > > wrote: > >> > >> > >> > >> On 2020/12/22 20:42, Fujii Masao wrote: > >>> > >>> > >>> On 2020/12/22 10:25, Masahiko Sawada wrote: > On Fr

Re: Deadlock between backend and recovery may not be detected

2020-12-23 Thread Fujii Masao
On 2020/12/23 19:28, Masahiko Sawada wrote: On Tue, Dec 22, 2020 at 11:58 PM Fujii Masao wrote: On 2020/12/22 20:42, Fujii Masao wrote: On 2020/12/22 10:25, Masahiko Sawada wrote: On Fri, Dec 18, 2020 at 6:36 PM Fujii Masao wrote: On 2020/12/17 2:15, Fujii Masao wrote: On 2020/

Re: Deadlock between backend and recovery may not be detected

2020-12-23 Thread Masahiko Sawada
On Tue, Dec 22, 2020 at 11:58 PM Fujii Masao wrote: > > > > On 2020/12/22 20:42, Fujii Masao wrote: > > > > > > On 2020/12/22 10:25, Masahiko Sawada wrote: > >> On Fri, Dec 18, 2020 at 6:36 PM Fujii Masao > >> wrote: > >>> > >>> > >>> > >>> On 2020/12/17 2:15, Fujii Masao wrote: > > >

Re: Deadlock between backend and recovery may not be detected

2020-12-22 Thread Fujii Masao
On 2020/12/22 20:42, Fujii Masao wrote: On 2020/12/22 10:25, Masahiko Sawada wrote: On Fri, Dec 18, 2020 at 6:36 PM Fujii Masao wrote: On 2020/12/17 2:15, Fujii Masao wrote: On 2020/12/16 23:28, Drouvot, Bertrand wrote: Hi, On 12/16/20 2:36 PM, Victor Yegorov wrote: *CAUTION*: Th

Re: Deadlock between backend and recovery may not be detected

2020-12-22 Thread Fujii Masao
On 2020/12/22 10:25, Masahiko Sawada wrote: On Fri, Dec 18, 2020 at 6:36 PM Fujii Masao wrote: On 2020/12/17 2:15, Fujii Masao wrote: On 2020/12/16 23:28, Drouvot, Bertrand wrote: Hi, On 12/16/20 2:36 PM, Victor Yegorov wrote: *CAUTION*: This email originated from outside of the or

Re: Deadlock between backend and recovery may not be detected

2020-12-22 Thread Fujii Masao
On 2020/12/19 1:43, Drouvot, Bertrand wrote: Hi, On 12/18/20 10:35 AM, Fujii Masao wrote: CAUTION: This email originated from outside of the organization. Do not click links or open attachments unless you can confirm the sender and know the content is safe. On 2020/12/17 2:15, Fujii Mas

Re: Deadlock between backend and recovery may not be detected

2020-12-21 Thread Masahiko Sawada
On Fri, Dec 18, 2020 at 6:36 PM Fujii Masao wrote: > > > > On 2020/12/17 2:15, Fujii Masao wrote: > > > > > > On 2020/12/16 23:28, Drouvot, Bertrand wrote: > >> Hi, > >> > >> On 12/16/20 2:36 PM, Victor Yegorov wrote: > >>> > >>> *CAUTION*: This email originated from outside of the organization. D

Re: Deadlock between backend and recovery may not be detected

2020-12-18 Thread Fujii Masao
On 2020/12/17 2:15, Fujii Masao wrote: On 2020/12/16 23:28, Drouvot, Bertrand wrote: Hi, On 12/16/20 2:36 PM, Victor Yegorov wrote: *CAUTION*: This email originated from outside of the organization. Do not click links or open attachments unless you can confirm the sender and know the co

Re: Deadlock between backend and recovery may not be detected

2020-12-16 Thread Fujii Masao
On 2020/12/16 23:28, Drouvot, Bertrand wrote: Hi, On 12/16/20 2:36 PM, Victor Yegorov wrote: *CAUTION*: This email originated from outside of the organization. Do not click links or open attachments unless you can confirm the sender and know the content is safe. ср, 16 дек. 2020 г. в 13:

Re: Deadlock between backend and recovery may not be detected

2020-12-16 Thread Victor Yegorov
ср, 16 дек. 2020 г. в 13:49, Fujii Masao : > After doing this procedure, you can see the startup process and backend > wait for the table lock each other, i.e., deadlock. But this deadlock > remains > even after deadlock_timeout passes. > > This seems a bug to me. > > > * Deadlocks involving the S

Deadlock between backend and recovery may not be detected

2020-12-16 Thread Fujii Masao
Hi, While reviewing the patch proposed at [1], I found that there is the case where deadlock that recovery conflict on lock is involved in may not be detected. This deadlock can happen between backends and the startup process, in the standby server. Please see the following procedure to reproduce