Обсуждение: pg_dump in stand alone backend
Hi, I would like to stop the postmaster every night and run vacuum pg_dump reindex in the stand alone backend. Vacuum and reindex seem to be quite easy, as I can setup a small script with both commands. But what about pg_dump. That seems "somewhat" more complex. TIA Ulrich
Hi, Citing Ulrich Wisser <ulrich.wisser@relevanttraffic.se>: > I would like to stop the postmaster every night and run > > vacuum > pg_dump > reindex > > in the stand alone backend. > > Vacuum and reindex seem to be quite easy, as I can setup a small script > with both commands. But what about pg_dump. That seems "somewhat" more > complex. What exactly is your problem about putting pg_dump in a (shell)script as well? In the simplest case you could use something like: #!/bin/sh psql --command vacuum your_database cd /somewhere/with/write/access pg_dump your_database > dump_`date %Y%m%d%H%M%S`.bak psql --command 'reindex whatever_you_want_to_reindex' your_database which will vacuum, dump to a file named dump_timewhendumpoccured.bak and reindex whatever_you_want_to_reindex. All kinds of stuff could be added (like mailing command output to you, loading the backup up to another machine for storage etc. etc.), but above script does basically what you were asking for. Regards, Daniel
Daniel Martini wrote: > Hi, > > Citing Ulrich Wisser <ulrich.wisser@relevanttraffic.se>: > >>I would like to stop the postmaster every night and run >> >>vacuum >>pg_dump >>reindex >> >>in the stand alone backend. >> >>Vacuum and reindex seem to be quite easy, as I can setup a small script >>with both commands. But what about pg_dump. That seems "somewhat" more >>complex. > > > What exactly is your problem about putting pg_dump in a (shell)script as > well? psql will not work without postmaster running. But if I let postmaster run some other tools will connect to it and I can't get exclusive locks on all tables for "vacuum full" and "reindex". With vacuum and reindex I can do #!/bin/bash postgres -d /var/lib/pgsql/data -U xxx -O -P mydb < SQL REINDEX DATABASE mydb REINDEX TABLE t1 FORCE REINDEX TABLE t2 ... REINDEX TABLE tn VACUUM FULL VERBOSE ANALYZE <<SQL which will speed up my database on a daily basis at least by factor 5. For security reasons I would also like to take a backup and I guess it will be much faster in the stand alone backend. Ulrich
Ulrich Wisser wrote: > Hi, > > I would like to stop the postmaster every night and run > > vacuum > pg_dump > reindex > > in the stand alone backend. > > Vacuum and reindex seem to be quite easy, as I can setup a small script > with both commands. But what about pg_dump. That seems "somewhat" more > complex. Explain what exactly you are trying to do, why do you have to stop the postmaster ? If you request is due only to forbid the access then you can replace the pg_hba.conf with a "void" one and replace it again at the end of operations. BTW vacuum, pg_dump, reindex are operations that can be performed with the server up and running. Regards Gaetano Mendola
Hi, >> I would like to stop the postmaster every night and run >> >> vacuum >> pg_dump >> reindex >> >> in the stand alone backend. >> >> Vacuum and reindex seem to be quite easy, as I can setup a small >> script with both commands. But what about pg_dump. That seems >> "somewhat" more complex. > > > Explain what exactly you are trying to do, why do you have to stop > the postmaster ? If you request is due only to forbid the access then > you can replace the pg_hba.conf with a "void" one and replace it again > at the end of operations. > > BTW vacuum, pg_dump, reindex are operations that can be performed > with the server up and running. on my database server I have a lot of scripts running. Some of them are not under my control and some of these don't really behave nice. So when my "maintainance period" starts they will still be running for hours and take exclusive locks on 100,000+ rows in various tables. Which means neither "vacuum full" nor "reindex" can get locks on these tables. Which stalls these calls and delays them to a time when my well behaving but heavy duty scripts start running again. Now these scripts will be delayed be "vacuum" or "reindex" and when I get to my desk the next morning the whole system is in overload. I need to to disconnect any other users and do "vacuum full verbose analyze" "reindex database" and reindex all tables. And for these I will stop the postmaster and run a stand alone backend. I figured that doing a nightly backup would be a good idea and running it in stand alone mode will speed up the process drastically. Ulrich
Ulrich Wisser wrote: >> >> Explain what exactly you are trying to do, why do you have to stop >> the postmaster ? If you request is due only to forbid the access then >> you can replace the pg_hba.conf with a "void" one and replace it again >> at the end of operations. [snip] > I need to to disconnect any other users and do "vacuum full verbose > analyze" "reindex database" and reindex all tables. And for these I will > stop the postmaster and run a stand alone backend. > > I figured that doing a nightly backup would be a good idea and running > it in stand alone mode will speed up the process drastically. Don't think it makes much difference, assuming you're the only one connected. I'd follow Gaetano's idea and have separate postgresql.conf/pg_hba.conf files. 1. stop PG 2. swap conf files 3. start PG 4. maintenance 5. stop PG 6. swap conf files back 7. start PG A separate postgresql.conf lets you have different sort_mem values etc. for your nightly maintenance too. -- Richard Huxton Archonet Ltd
Hi, Citing Ulrich Wisser <ulrich.wisser@relevanttraffic.se>: > on my database server I have a lot of scripts running. Some of them are > not under my control and some of these don't really behave nice. So when > my "maintainance period" starts they will still be running for hours and > take exclusive locks on 100,000+ rows in various tables. Which means > neither "vacuum full" nor "reindex" can get locks on these tables. Which > stalls these calls and delays them to a time when my well behaving but > heavy duty scripts start running again. Now these scripts will be > delayed be "vacuum" or "reindex" and when I get to my desk the next > morning the whole system is in overload. > > I need to to disconnect any other users and do "vacuum full verbose > analyze" "reindex database" and reindex all tables. And for these I will > stop the postmaster and run a stand alone backend. How about stopping the postmaster, running vacuum and reindex in the standalone backend, then starting the postmaster with a different pg_hba.conf and optionally setting a few options from postgresql.conf from the commandline (e.g. MAX_CONNECTIONS, SUPERUSER_RESERVED_CONNECTIONS) to limit access to the server to the user doing the dump and then doing the dump. When it is finished, you could restart the postmaster with the proper runtime environment in place. Would this work? > I figured that doing a nightly backup would be a good idea and running > it in stand alone mode will speed up the process drastically. It seems, we can't try if it is really drastically faster than just starting the postmaster with restricted access and then doing the dump. Another option would be to look at the source to determine what pg_dump actually does and write a sql-script to come as close as possible to that and postprocess the output of feeding this to the standalone backend to produce again sql which could be loaded by psql. Probably a lot of work, but perhaps doable if you only care about the actual data and not about triggers, functions, access permissions etc. etc. Regards, Daniel
Ulrich Wisser <ulrich.wisser@relevanttraffic.se> writes:
> I need to to disconnect any other users and do "vacuum full verbose
> analyze" "reindex database" and reindex all tables. And for these I will
> stop the postmaster and run a stand alone backend.
I think the real problem here is stone-age maintenance procedures ;-)
You shouldn't need to do vacuum full on a regular basis, and you
shouldn't need to do reindexing on a regular basis either. Update
to 7.4, if you aren't using it already, and replace these procedures
by plain vacuums run often enough to keep the DB from bloating (a look
at your FSM parameters would be advisable too).
> I figured that doing a nightly backup would be a good idea and running
> it in stand alone mode will speed up the process drastically.
No it won't.
regards, tom lane