| From: | "Andrey M(dot) Borodin" <x4mmm(at)yandex-team(dot)ru> |
|---|---|
| To: | Parag Paul <parag(dot)paul(at)gmail(dot)com> |
| Cc: | Robert Haas <robertmhaas(at)gmail(dot)com>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Andres Freund <andres(at)anarazel(dot)de>, pgsql-hackers(at)lists(dot)postgresql(dot)org |
| Subject: | Re: Issue with the PRNG used by Postgres |
| Date: | 2024-04-11 05:52:43 |
| Message-ID: | D3091CDA-B0BB-4EF7-8BAA-63519248B549@yandex-team.ru |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
> On 10 Apr 2024, at 21:48, Parag Paul <parag(dot)paul(at)gmail(dot)com> wrote:
>
> Yes, the probability of this happening is astronomical, but in production with 128 core servers with 7000 max_connections, with petabyte scale data, this did repro 2 times in the last month. We had to move to a local approach to manager our ratelimiting counters.
FWIW we observed such failure on this [0] LWLock two times too. Both cases were recent (February).
We have ~15k clusters with 8MTPS, so it’s kind of infrequent, but not astronomic. We decided to remove that lock.
Best regards, Andrey Borodin.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Ashutosh Bapat | 2024-04-11 06:37:09 | apply_scanjoin_target_to_paths and partitionwise join |
| Previous Message | Pavel Stehule | 2024-04-11 05:34:08 | Re: Schema variables - new implementation for Postgres 15 |