mirrors/postgres
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Revert ill-considered change in pg_resetwal output.

Browse Source 
Commit 31dcfae83 changed one pg_resetwal output string, and a
corresponding test in pg_upgrade, without sufficient thought for
the consequences.  We can't change that output without creating
hazards for cross-version upgrades, since pg_upgrade needs to be able
to read the output of several different versions of pg_resetwal.
There may well be external tools with the same requirement.

For the moment, just revert those two changes.  What we really
ought to do here is have a separate, stable, easily machine-readable
output format for pg_resetwal and pg_controldata, as proposed
years ago by Alvaro.  Once that's in place and tools no longer
need to depend on the exact spelling of the human-readable output,
we can put back this change.

Discussion: https://postgr.es/m/fbea8c6f-415a-bad9-c3de-969c40d08a84@dunslane.net
This commit is contained in:
Tom Lane 2022-09-15 10:58:03 -04:00
parent b4f584f9d2
commit 918de37652
2 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/bin/pg_resetwal/pg_resetwal.c
View File

@ -788,7 +788,7 @@ PrintNewControlValues(void)
XLogFileName(fname, ControlFile.checkPointCopy.ThisTimeLineID,
newXlogSegNo, WalSegSz);
printf(_("First WAL segment after reset: %s\n"), fname);
printf(_("First log segment after reset: %s\n"), fname);
if (set_mxid != 0)
{

2
src/bin/pg_upgrade/controldata.c
View File

@ -348,7 +348,7 @@ get_control_data(ClusterInfo *cluster, bool live_check)
cluster->controldata.chkpnt_nxtmxoff = str2uint(p);
got_mxoff = true;
}
else if ((p = strstr(bufin, "First WAL segment after reset:")) != NULL)
else if ((p = strstr(bufin, "First log segment after reset:")) != NULL)
{
/* Skip the colon and any whitespace after it */
p = strchr(p, ':');