Reputation
1,882
Top tag
Next privilege 2,000 Rep.
Edit questions and answers
Badges
5 17
Newest
 Curious
Impact
~94k people reached

2d
comment What to do with WAL files for Postgres Slave reset
If you have a WAL achive, you can try restore_command as Craig Ringer suggests. The pg_basebackup creates an entirely new slave in an empty directory.
2d
answered What to do with WAL files for Postgres Slave reset
2d
comment What to do with WAL files for Postgres Slave reset
Could you post more of the logs, from where it starts to recover to the error? If it's a lot of text, you could link to pastebin.com
2d
comment What to do with WAL files for Postgres Slave reset
What happens if you start the slave? Why can it not resume at the point where it stopped?
Apr
22
comment Anything like a hash for Postgresql databases?
We've got scripts that rely on exactly that and they've been working for years. Of course we might just be lucky. Do you have any sources to back your claim that the order of things in pg_dump is arbitrary?
Apr
22
comment PostgreSQL DB freezes with large inserts
Are you sure Postgres is causing the memory usage growth? Usually Postgres restrains itself to a small subset of memory (the shared_buffers setting)
Apr
22
answered Anything like a hash for Postgresql databases?
Apr
22
revised Performance hickups after Postgres 9.4 upgrade
edited title
Apr
21
revised Performance hickups after Postgres 9.4 upgrade
added 111 characters in body
Apr
21
comment Performance hickups after Postgres 9.4 upgrade
Turns out the extension locks were just a manifestation of a problem with transparent huge pages. From what I read, setting a manual fillfactor only helps with HOT (heap only tuple) workloads. That's rather uncommon.
Apr
21
accepted Performance hickups after Postgres 9.4 upgrade
Apr
21
answered Performance hickups after Postgres 9.4 upgrade
Apr
17
comment Performance hickups after Postgres 9.4 upgrade
Thanks. An extension doesn't seem like a CPU intensive operation... any idea on how an extension could cause high CPU load?
Apr
17
awarded  Curious
Apr
16
asked Performance hickups after Postgres 9.4 upgrade
Mar
19
comment Count data using multiple date ranges
Right, in this case it is. But when you read this query in a year or so, you'll have to double-check the type of the date column
Mar
18
comment Count data using multiple date ranges
@a_horse_with_no_name: Updated to use clicks, though Valentin seems to have figured that out by himself :)
Mar
18
comment Count data using multiple date ranges
Looks much better, filter is great syntactic sugar! The query assumes date is a column of type date and not timestamp. That assumption has burned me at times
Mar
18
revised Count data using multiple date ranges
added 16 characters in body
Mar
18
revised Count data using multiple date ranges
added 8 characters in body