public inbox for [email protected]help / color / mirror / Atom feed
Postgres Logical Replication - how to see what subscriber is doing with received data? 5+ messages / 3 participants [nested] [flat]
* Postgres Logical Replication - how to see what subscriber is doing with received data? @ 2024-08-29 01:58 Michael Jaskiewicz <[email protected]> 0 siblings, 1 reply; 5+ messages in thread From: Michael Jaskiewicz @ 2024-08-29 01:58 UTC (permalink / raw) To: pgsql-general I've got two Postgres 13 databases on AWS RDS. * One is a master, the other a slave using logical replication. * Replication has fallen behind by about 350Gb. * The slave was maxed out in terms of CPU for the past four days because of some jobs that were ongoing so I'm not sure what logical replication was able to replicate during that time. * I killed those jobs and now CPU on the master and slave are both low. * I look at the subscriber via `select * from pg_stat_subscription;` and see that latest_end_lsn is advancing albeit very slowly. * The publisher says write/flush/replay lags are all 13 minutes behind but it's been like that for most of the day. * I see no errors in the logs on either the publisher or subscriber outside of some simple SQL errors that users have been making. * CloudWatch reports low CPU utilization, low I/O, and low network. Is there anything I can do here? Previously I set wal_receiver_timeout timeout to 0 because I had replication issues, and that helped things. I wish I had some visibility here to get any kind of confidence that it's going to pull through, but other than these lsn values and database logs, I'm not sure what to check. Sincerely, mj ^ permalink raw reply [nested|flat] 5+ messages in thread
* Re: Postgres Logical Replication - how to see what subscriber is doing with received data? @ 2024-09-01 16:22 Shaheed Haque <[email protected]> parent: Michael Jaskiewicz <[email protected]> 0 siblings, 1 reply; 5+ messages in thread From: Shaheed Haque @ 2024-09-01 16:22 UTC (permalink / raw) To: Michael Jaskiewicz <[email protected]>; +Cc: pgsql-general Since nobody more knowledgeable has replied... I'm very interested in this area and still surprised that there is no official/convenient/standard way to approach this (see https://www.postgresql.org/message-id/CAHAc2jdAHvp7tFZBP37awcth%3DT3h5WXCN9KjZOvuTNJaAAC_hg%40mail.g... ). Based partly on that thread, I ended up with a script that connects to both ends of the replication, and basically loops while comparing the counts in each table. On Fri, 30 Aug 2024, 12:38 Michael Jaskiewicz, <[email protected]> wrote: > I've got two Postgres 13 databases on AWS RDS. > > - One is a master, the other a slave using logical replication. > - Replication has fallen behind by about 350Gb. > - The slave was maxed out in terms of CPU for the past four days > because of some jobs that were ongoing so I'm not sure what logical > replication was able to replicate during that time. > - I killed those jobs and now CPU on the master and slave are both low. > - I look at the subscriber via `select * from pg_stat_subscription;` > and see that latest_end_lsn is advancing albeit very slowly. > - The publisher says write/flush/replay lags are all 13 minutes behind > but it's been like that for most of the day. > - I see no errors in the logs on either the publisher or subscriber > outside of some simple SQL errors that users have been making. > - CloudWatch reports low CPU utilization, low I/O, and low network. > > > > Is there anything I can do here? Previously I set wal_receiver_timeout > timeout to 0 because I had replication issues, and that helped things. I > wish I had *some* visibility here to get any kind of confidence that it's > going to pull through, but other than these lsn values and database logs, > I'm not sure what to check. > > > > Sincerely, > > mj > ^ permalink raw reply [nested|flat] 5+ messages in thread
* Re: Postgres Logical Replication - how to see what subscriber is doing with received data? @ 2024-09-02 06:08 Muhammad Ikram <[email protected]> parent: Shaheed Haque <[email protected]> 0 siblings, 1 reply; 5+ messages in thread From: Muhammad Ikram @ 2024-09-02 06:08 UTC (permalink / raw) To: Shaheed Haque <[email protected]>; +Cc: Michael Jaskiewicz <[email protected]>; pgsql-general Hi Shaheed, Maybe these considerations could help you or give any hint to the problem ? Check if wal_receiver_timeout being set to 0 could potentially cause issues, like not detecting network issues quickly enough. Consider re-evaluating this setting if you see connection issues. If you notice that some data is missing on subscriber then could you increase max_slot_wal_keep_size on publisher so that WALs are not deleted until they are applied on subscriber. Do you have flexibility to increase max_worker_processes and max_logical_replication_workers, work_mem and maintenance_work_mem on subscriber (In case bottleneck exists on subscriber) If there's significant lag, consider whether it might be more efficient to drop the subscription and re-initialize it from scratch using a new base backup, depending on the data volume and how long it might take for the existing replication to catch up. Regards, Muhammad Ikram On Sun, Sep 1, 2024 at 9:22 PM Shaheed Haque <[email protected]> wrote: > Since nobody more knowledgeable has replied... > > I'm very interested in this area and still surprised that there is no > official/convenient/standard way to approach this (see > https://www.postgresql.org/message-id/CAHAc2jdAHvp7tFZBP37awcth%3DT3h5WXCN9KjZOvuTNJaAAC_hg%40mail.g... > ). > > Based partly on that thread, I ended up with a script that connects to > both ends of the replication, and basically loops while comparing the > counts in each table. > > On Fri, 30 Aug 2024, 12:38 Michael Jaskiewicz, <[email protected]> > wrote: > >> I've got two Postgres 13 databases on AWS RDS. >> >> - One is a master, the other a slave using logical replication. >> - Replication has fallen behind by about 350Gb. >> - The slave was maxed out in terms of CPU for the past four days >> because of some jobs that were ongoing so I'm not sure what logical >> replication was able to replicate during that time. >> - I killed those jobs and now CPU on the master and slave are both >> low. >> - I look at the subscriber via `select * from pg_stat_subscription;` >> and see that latest_end_lsn is advancing albeit very slowly. >> - The publisher says write/flush/replay lags are all 13 minutes >> behind but it's been like that for most of the day. >> - I see no errors in the logs on either the publisher or subscriber >> outside of some simple SQL errors that users have been making. >> - CloudWatch reports low CPU utilization, low I/O, and low network. >> >> >> >> Is there anything I can do here? Previously I set wal_receiver_timeout >> timeout to 0 because I had replication issues, and that helped things. I >> wish I had *some* visibility here to get any kind of confidence that >> it's going to pull through, but other than these lsn values and database >> logs, I'm not sure what to check. >> >> >> >> Sincerely, >> >> mj >> > -- Muhammad Ikram ^ permalink raw reply [nested|flat] 5+ messages in thread
* Re: Postgres Logical Replication - how to see what subscriber is doing with received data? @ 2024-09-02 07:42 Shaheed Haque <[email protected]> parent: Muhammad Ikram <[email protected]> 0 siblings, 1 reply; 5+ messages in thread From: Shaheed Haque @ 2024-09-02 07:42 UTC (permalink / raw) To: Muhammad Ikram <[email protected]>; +Cc: Michael Jaskiewicz <[email protected]>; pgsql-general Hi Muhammad, On Mon, 2 Sep 2024, 07:08 Muhammad Ikram, <[email protected]> wrote: > Hi Shaheed, > > Maybe these considerations could help you or give any hint to the problem ? > > > Check if wal_receiver_timeout being set to 0 could potentially cause > issues, like not detecting network issues quickly enough. Consider > re-evaluating this setting if you see connection issues. > > If you notice that some data is missing on subscriber then could you > increase max_slot_wal_keep_size on publisher so that WALs are not deleted > until they are applied on subscriber. > > Do you have flexibility to increase max_worker_processes and > max_logical_replication_workers, work_mem and maintenance_work_mem on > subscriber (In case bottleneck exists on subscriber) > > If there's significant lag, consider whether it might be more efficient to > drop the subscription and re-initialize it from scratch using a new base > backup, depending on the data volume and how long it might take for the > existing replication to catch up. > Thanks for the kind hints, I'll certainly look into those. My main interest however was with the "visibility" question, i.e. to get an understanding of the gap between the two ends of a replication slot, ideally in human terms (e.g. tables x records). I understand the difficulties of trying to produce a meaningful metric that spans two (or more) systems but let's be honest, trying to diagnose which knobs to tweak (whether in application, PG, the OS or the network) is basically black magic when all we really have is a pair of opaque LSNs. > > Regards, > Muhammad Ikram > > > On Sun, Sep 1, 2024 at 9:22 PM Shaheed Haque <[email protected]> > wrote: > >> Since nobody more knowledgeable has replied... >> >> I'm very interested in this area and still surprised that there is no >> official/convenient/standard way to approach this (see >> https://www.postgresql.org/message-id/CAHAc2jdAHvp7tFZBP37awcth%3DT3h5WXCN9KjZOvuTNJaAAC_hg%40mail.g... >> ). >> >> Based partly on that thread, I ended up with a script that connects to >> both ends of the replication, and basically loops while comparing the >> counts in each table. >> >> On Fri, 30 Aug 2024, 12:38 Michael Jaskiewicz, <[email protected]> >> wrote: >> >>> I've got two Postgres 13 databases on AWS RDS. >>> >>> - One is a master, the other a slave using logical replication. >>> - Replication has fallen behind by about 350Gb. >>> - The slave was maxed out in terms of CPU for the past four days >>> because of some jobs that were ongoing so I'm not sure what logical >>> replication was able to replicate during that time. >>> - I killed those jobs and now CPU on the master and slave are both >>> low. >>> - I look at the subscriber via `select * from pg_stat_subscription;` >>> and see that latest_end_lsn is advancing albeit very slowly. >>> - The publisher says write/flush/replay lags are all 13 minutes >>> behind but it's been like that for most of the day. >>> - I see no errors in the logs on either the publisher or subscriber >>> outside of some simple SQL errors that users have been making. >>> - CloudWatch reports low CPU utilization, low I/O, and low network. >>> >>> >>> >>> Is there anything I can do here? Previously I set wal_receiver_timeout >>> timeout to 0 because I had replication issues, and that helped things. I >>> wish I had *some* visibility here to get any kind of confidence that >>> it's going to pull through, but other than these lsn values and database >>> logs, I'm not sure what to check. >>> >>> >>> >>> Sincerely, >>> >>> mj >>> >> > > -- > Muhammad Ikram > > ^ permalink raw reply [nested|flat] 5+ messages in thread
* Re: Postgres Logical Replication - how to see what subscriber is doing with received data? @ 2024-09-02 08:45 Muhammad Ikram <[email protected]> parent: Shaheed Haque <[email protected]> 0 siblings, 0 replies; 5+ messages in thread From: Muhammad Ikram @ 2024-09-02 08:45 UTC (permalink / raw) To: Shaheed Haque <[email protected]>; +Cc: Michael Jaskiewicz <[email protected]>; pgsql-general Hi Shaheed, I think you must have already analyzed the outcome of queries on pg_replication_slots, pg_current_wal_lsn(), pg_stat_subscription etc. I could find a query SELECT pg_size_pretty(pg_wal_lsn_diff('<publisher_restart_lsn>', '<subscriber_replayed_lsn>')); As a side note if you want to see what has been applied to subscribers vs what exists on publisher then here is something from my previous experience. We used to have a Data Validation tool for checking tables/rows across publisher/subscriber. We also used pg_dump for another tool that was meant for making copies of schemas. Regards, Muhammad Ikram On Mon, Sep 2, 2024 at 12:42 PM Shaheed Haque <[email protected]> wrote: > Hi Muhammad, > > On Mon, 2 Sep 2024, 07:08 Muhammad Ikram, <[email protected]> wrote: > >> Hi Shaheed, >> >> Maybe these considerations could help you or give any hint to the problem >> ? >> >> >> Check if wal_receiver_timeout being set to 0 could potentially cause >> issues, like not detecting network issues quickly enough. Consider >> re-evaluating this setting if you see connection issues. >> >> If you notice that some data is missing on subscriber then could you >> increase max_slot_wal_keep_size on publisher so that WALs are not deleted >> until they are applied on subscriber. >> >> Do you have flexibility to increase max_worker_processes and >> max_logical_replication_workers, work_mem and maintenance_work_mem on >> subscriber (In case bottleneck exists on subscriber) >> >> If there's significant lag, consider whether it might be more efficient >> to drop the subscription and re-initialize it from scratch using a new base >> backup, depending on the data volume and how long it might take for the >> existing replication to catch up. >> > > Thanks for the kind hints, I'll certainly look into those. > > My main interest however was with the "visibility" question, i.e. to get > an understanding of the gap between the two ends of a replication slot, > ideally in human terms (e.g. tables x records). > > I understand the difficulties of trying to produce a meaningful metric > that spans two (or more) systems but let's be honest, trying to diagnose > which knobs to tweak (whether in application, PG, the OS or the network) is > basically black magic when all we really have is a pair of opaque LSNs. > > > > >> >> Regards, >> Muhammad Ikram >> >> >> On Sun, Sep 1, 2024 at 9:22 PM Shaheed Haque <[email protected]> >> wrote: >> >>> Since nobody more knowledgeable has replied... >>> >>> I'm very interested in this area and still surprised that there is no >>> official/convenient/standard way to approach this (see >>> https://www.postgresql.org/message-id/CAHAc2jdAHvp7tFZBP37awcth%3DT3h5WXCN9KjZOvuTNJaAAC_hg%40mail.g... >>> ). >>> >>> Based partly on that thread, I ended up with a script that connects to >>> both ends of the replication, and basically loops while comparing the >>> counts in each table. >>> >>> On Fri, 30 Aug 2024, 12:38 Michael Jaskiewicz, <[email protected]> >>> wrote: >>> >>>> I've got two Postgres 13 databases on AWS RDS. >>>> >>>> - One is a master, the other a slave using logical replication. >>>> - Replication has fallen behind by about 350Gb. >>>> - The slave was maxed out in terms of CPU for the past four days >>>> because of some jobs that were ongoing so I'm not sure what logical >>>> replication was able to replicate during that time. >>>> - I killed those jobs and now CPU on the master and slave are both >>>> low. >>>> - I look at the subscriber via `select * from >>>> pg_stat_subscription;` and see that latest_end_lsn is advancing albeit very >>>> slowly. >>>> - The publisher says write/flush/replay lags are all 13 minutes >>>> behind but it's been like that for most of the day. >>>> - I see no errors in the logs on either the publisher or subscriber >>>> outside of some simple SQL errors that users have been making. >>>> - CloudWatch reports low CPU utilization, low I/O, and low network. >>>> >>>> >>>> >>>> Is there anything I can do here? Previously I set wal_receiver_timeout >>>> timeout to 0 because I had replication issues, and that helped things. I >>>> wish I had *some* visibility here to get any kind of confidence that >>>> it's going to pull through, but other than these lsn values and database >>>> logs, I'm not sure what to check. >>>> >>>> >>>> >>>> Sincerely, >>>> >>>> mj >>>> >>> >> >> -- >> Muhammad Ikram >> >> -- Muhammad Ikram ^ permalink raw reply [nested|flat] 5+ messages in thread
end of thread, other threads:[~2024-09-02 08:45 UTC | newest] Thread overview: 5+ messages (download: mbox mbox.gz follow: Atom feed) -- links below jump to the message on this page -- 2024-08-29 01:58 Postgres Logical Replication - how to see what subscriber is doing with received data? Michael Jaskiewicz <[email protected]> 2024-09-01 16:22 ` Shaheed Haque <[email protected]> 2024-09-02 06:08 ` Muhammad Ikram <[email protected]> 2024-09-02 07:42 ` Shaheed Haque <[email protected]> 2024-09-02 08:45 ` Muhammad Ikram <[email protected]>
This inbox is served by agora; see mirroring instructions for how to clone and mirror all data and code used for this inbox