On Mon, 2019-05-13 at 23:39 -0700, David G. Johnston wrote:
> On Mon, May 13, 2019 at 11:24 PM magodo wrote:
> > I found when running command like `# echo "xxx" | psql postgres
> >
> > postgres`, the return code is always 0 even though the command
> > (&quo
Hello,
I found when running command like `# echo "xxx" | psql postgres
postgres`, the return code is always 0 even though the command ("xxx")
here is of invalid syntax. While the `psql -c` way handled exit code
correctly.
In the meanwhile, mysql client handled both cases correctly.
---
Zhaotin
ote. Is there anything I need to wait before
I rewind the old primary against this promoted cluster?
Thank you in advance!
---
magodo
xlogdump -r
Transaction`, and try to find the first least earliest WAL against the
specified time. This is a linear search, which has complexity of O(n).
So if you want to recover to that point of time, how do you choose the
timeline?
---
Magodo
restart it,
which will end up with standby's LSN newer than the current flushed one
on primary.
My solution to this problem is to do a restart also on primary. I
wonder if this is a proper way.
Thank you in advance!
---
Magodo
On Thu, 2018-10-11 at 06:35 +0200, Laurenz Albe wrote:
> magodo wrote:
> > OK... Just take another example:
> >
> > A B
> > BASE-+-+--o1 (recover to
> > A) 1
> > | | C
> &
On Wed, 2018-10-10 at 18:21 +0200, Laurenz Albe wrote:
> magodo wrote:
> > > By default, recovery will stay on the timeline where it started.
> > > If you want to go to timeline 2 or 3, you have to specify
> > > recovery_target_timeline.
> > >
> &g
On Wed, 2018-10-03 at 08:06 +0200, Laurenz Albe wrote:
> magodo wrote:
> > I read the document about recovery configuration, it says:
>
> [...]
>
> > Therefore, suppose I am going to recover to a point of some child
> > timeline, to identify the p
to parse each wal and compare with the specified datetime
in a binary-search way, but it is based on the fact that every COMMIT
time in wal is sequential, but it looks like not the case..
Hence I wonder what is the best practice for this task?
Thank you in advance!
---
Magodo
On Wed, 2018-10-10 at 11:24 +0800, magodo wrote:
> Hello,
>
> I'm figuring out how to do PITR on a HA setup for postgres 9.6. I
> almost manage to do it, while I faced following issue:
>
>
> [1. create basebackup]
>
> [2. insert 2 rows]
>
>
3:09:57.119 UTC > DEBUG: archived transaction log file
"00020005"
< 2018-10-10 03:09:57.759 UTC > DEBUG: forked new backend, pid=254
socket=11
< 2018-10-10 03:09:57.771 UTC > DEBUG: server process (PID 254) exited
with exit code 0
And I have to execute `pg_xlog_replay_resume()` on the new primary so
that it will branch out to new timeline 3 and work normally...
Would anyone please tell me what's going on here? How can I avoid doing
a manual resume?
---
Magodo
On Tue, 2018-10-09 at 16:55 +0900, Michael Paquier wrote:
> On Tue, Oct 09, 2018 at 03:26:35PM +0800, magodo wrote:
> > Yes, but does this differ whether I'm archiving a general WAL or
> > archiving the backup history? I mean if user doesn't handle
> > duplicat
On Tue, 2018-10-09 at 12:53 +0900, Michael Paquier wrote:
> On Tue, Oct 09, 2018 at 11:45:50AM +0800, magodo wrote:
> > Yet, I am still not so clear how does the bloat of pg_wal happen?
> > Do
> > you mean pg_wal will be filled up by many .backup(s)?
>
> If your archi
On Tue, 2018-10-09 at 11:45 +0900, Michael Paquier wrote:
> (Please do not forget to add the community mailing list in CC.)
>
> On Tue, Oct 09, 2018 at 10:33:56AM +0800, magodo wrote:
> > Since the backup history aims to aid administrator to identify the
> > point from whi
primary.
Is this as expected? If so, in which case should I do backup on primary
and in which case should I do it on standby?
Thank you in advance!
Magodo
I read the document about recovery configuration, it says:
recovery_target_timeline (string)
...
Other than that you only need to set this parameter in complex re-
recovery situations, where you need to return to a state that itself
was reached after a point-in-time recovery. Se
Hello,
I'm working on a high availability project for PG(v9.6).
Suppose I have a priamry-standby setup, I need to do promote once
primary is down, but only when standby is as new as primary (before it
is down). Therefore, I need a way to know the wal replay location on
standby.
I tried to set
17 matches
Mail list logo
