| From: | Kyotaro Horiguchi <horikyota(dot)ntt(at)gmail(dot)com> |
|---|---|
| To: | pgsql-hackers(at)lists(dot)postgresql(dot)org |
| Subject: | Strange error message in xlog.c |
| Date: | 2019-06-17 08:31:03 |
| Message-ID: | CAKPRHzJHSEgdon_Dgqe5bE33wiM2_mFVijywXTDoSVNMsb_qUQ@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hello,
I had an trouble report that the reporter had the following error
messages.
FATAL: XX000: requested timeline 175 is not a child of this server's history
DETAIL: Latest checkpoint is at 1A/D6000028 on timeline 172, but in
the history of the requested timeline, the server forked off from that
timeline at 1C/29074DB8.
This message doesn't make sense. Perhaps timeline 172 started
after 1A/D6000028 instead.
The attached patch makes the error messages for both cases make sense.
FATAL: requested timeline 4 is not a child of this server's history
DETAIL: Latest checkpoint is at 0/3000060 on timeline 2, but in the
history of the requested timeline, the server forked off from that
timeline at 0/22000A0.
FATAL: requested timeline 4 is not a child of this server's history
DETAIL: Latest checkpoint is at 0/3000060 on timeline 2, but in the
history of the requested timeline, the server entered that timeline at
0/40000A0.
Intentional corruption of timeline-history is required to
exercise this. Do we need to do that regression test?
regards.
--
Kyotaro Horiguchi
NTT Open Source Software Center
| Attachment | Content-Type | Size |
|---|---|---|
| 0001-Fix-error-message-for-timeline-history-mismatch.patch | application/octet-stream | 4.0 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Alexander Korotkov | 2019-06-17 08:36:12 | Re: SQL/JSON path issues/questions |
| Previous Message | Jamison, Kirk | 2019-06-17 08:01:04 | RE: [PATCH] Speedup truncates of relation forks |