[Bucardo-general] Source DB error when adding the sync
Lucas Possamai
drum.lucas at gmail.com
Sun Apr 18 10:49:58 UTC 2021
Hi Andreas, thanks for your email.
Hmm... that sucks! :(
Did you perform the same/similar steps as I did? Just wanna make sure
they're correct.
Cheers,
Lucas
On Sat, 17 Apr 2021 at 23:49, Andreas Wahler <andreas.wahler at wamtek.net>
wrote:
> -hi Lucas,
>
> I faced these locks during adding a new sync to bucardo, too. During this
> step it helps a lot if no / less traffic is on the database.
>
> Best,
> Andreas
> Am 17.04.2021 um 13:08 schrieb Lucas Possamai:
>
> Anyone able to help with my previous email, please?
>
> I've just tried the process again, and I got into the same outcome. The
> steps I followed are below:
>
> ---
>
> 1. Created bucardo username on source and target dbs
> 2. because I've done this before, I've done some cleanup before
> repeating the process
> 1. bucardo remove sync test_sync
> 2. bucardo remove dbgroup test_db_group
> 3. bucardo remove relgroup test_db_group
> 4. bucardo remove all tables
> 5. bucardo remove all sequences
> 6. bucardo remove db source_db_ec2
> 7. bucardo remove target_db_rds
> 8. AS POSTGRES
> 1. drop schema bucardo cascade;
> 3.
>
> bucardo add db source_db_ec2 dbname=test_db host=
> pgsql-source.example.com user=bucardouser pass=password
> 4.
>
> bucardo add db target_db_rds dbname=test_db host=
> pgsql-target-rds.example.com user=bucardo pass=password
> 5.
>
> bucardo add table public.* db=source_db_ec2 relgroup=test_db_group
> 6.
>
> bucardo add table junk.* db=source_db_ec2 relgroup=test_db_group
> 7.
>
> bucardo add sequence public.* db=source_db_ec2 relgroup=test_db_group
> 8.
>
> bucardo add sequence junk.* db=source_db_ec2 relgroup=test_db_group
> 9.
>
> bucardo add dbgroup test_db_group source_db_ec2:source
> target_db_rds:target
> 10.
>
> bucardo add sync test_sync relgroup=test_db_group dbs=test_db_group
> autokick=1
> 11.
>
> bucardo validate test_sync
> 12.
>
> bucardo start
>
> When checking the status of the sync
>
> Last good : Apr 17, 2021 10:43:25 (time to run: 1s)
>
> Rows deleted/inserted : 0 / 0
>
> Sync name : test_sync
>
> Current state : Good
>
> Source relgroup/database : test_db_group / source_db_ec2
>
> Tables in sync : 209
>
> Status : Active
>
> Check time : None
>
> Overdue time : 00:00:00
>
> Expired time : 00:00:00
>
> Stayalive/Kidsalive : Yes / Yes
>
> Rebuild index : No
>
> Autokick : Yes
>
> Onetimecopy : No
>
> Post-copy analyze : Yes
>
> Last error: :
>
> ======================================================================
>
> ... everything looks normal
>
> Bucardo logs:
> (371540) [Sat Apr 17 10:42:26 2021] MCP Active syncs: 1
> (371540) [Sat Apr 17 10:42:26 2021] MCP Entering main loop
> (371542) [Sat Apr 17 10:42:26 2021] VAC New VAC daemon. PID=371542
> (371542) [Sat Apr 17 10:42:26 2021] VAC Connected to database
> "source_db_ec2" with backend PID of 81591
> (371540) [Sat Apr 17 10:42:26 2021] MCP Created VAC 371542
> (371544) [Sat Apr 17 10:42:26 2021] CTL New controller for sync
> "test_sync". Relgroup is "test_db_group", dbs is "test_db_group". PID=371544
> (371544) [Sat Apr 17 10:42:26 2021] CTL stayalive: 1 checksecs: 0
> kicked: 1
> (371544) [Sat Apr 17 10:42:26 2021] CTL kidsalive: 1 onetimecopy: 0
> lifetimesecs: 0 (NULL) maxkicks: 0
> (371544) [Sat Apr 17 10:42:26 2021] CTL Database "source_db_ec2" backend
> PID: 81592
> (371544) [Sat Apr 17 10:42:26 2021] CTL Database "target_db_rds" backend
> PID: 15208
> (371540) [Sat Apr 17 10:42:26 2021] MCP Created controller 371544 for sync
> "test_sync". Kick is 1
> (371546) [Sat Apr 17 10:42:27 2021] KID (test_sync) New kid, sync
> "test_sync" alive=1 Parent=371544 PID=371546 kicked=1
>
> Then, my DB servers stop processing queries from the application. I
> couldn't find anything on the PG logs.
>
> Can someone please tell me if the above steps are correct?
>
> Cheers
> ---
>
>
> ---
> Kind regards,
>
> *Lucas Possamai*
> lpossamai.me
>
>
> On Tue, 13 Apr 2021 at 10:31, Lucas Possamai <drum.lucas at gmail.com> wrote:
>
>> Hi all,
>>
>> When executing the command below, my source db collapsed. Looked like
>> lots of locks and waits were happening.
>>
>> bucardo add sync test_sync relgroup=test_db_group dbs=test_db_group
>> autokick=0
>>
>> The steps I'm doing are:
>>
>> - PG DUMP the database from source
>> - Create roles and apply database permissions to target
>> - Import the Database dump to the target DB (RDS)
>> - Enable session_replication_role on Target
>> - Bucardo add source DB
>> - Bucardo add target DB
>> - bucardo add tables
>> - bucardo add table public.* db=source_db relgroup=test_db_group
>> - bucardo add sequences
>> - bucardo add sequence public.* db=source_db relgroup=test_db_group
>> - bucardo add dbgroup
>> - bucardo add dbgroup test_db_group source_db:source
>> target_db:target
>> - bucardo add sync
>> - bucardo add sync test_sync relgroup=test_db_group
>> dbs=test_db_group autokick=0
>>
>>
>> Are those locks expected when adding the sync on the source db? Any other
>> recommendations?
>>
>> Source DB is PostgreSQL 9.2
>> Target DB is PostgreSQL 12.4
>> Bucardo version is: 5.6.0
>>
>> Cheers,
>> ---
>> Kind regards,
>>
>> *Lucas Possamai*
>> lpossamai.me
>>
>
> _______________________________________________
> Bucardo-general mailing listBucardo-general at bucardo.orghttps://bucardo.or
>
>
> _______________________________________________
> Bucardo-general mailing list
> Bucardo-general at bucardo.org
> https://bucardo.org/mailman/listinfo/bucardo-general
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://bucardo.org/pipermail/bucardo-general/attachments/20210418/4ae2b8c1/attachment-0001.htm>
More information about the Bucardo-general
mailing list