Change in ovirt-engine[master]: core: using pg_dump --column-inserts ...(#858643)
Yair Zaslavsky has submitted this change and it was merged.
Change subject: core: using pg_dump --column-inserts ...(#858643)
......................................................................
core: using pg_dump --column-inserts ...(#858643)
using pg_dump --column-inserts slows down the restore process
The backup script is calling internally the postgres pg_dump utility
that actually performs the backup.
The backup script output is a SQL file.
This file can have 2 formats for data inserts
1) COPY (per table) - efficient and fast, all data is inserted or none
2) INSERT (per table row) - slow and problematic in large databases,
part of table rows may be inserted since a failure in a row does not
skip the other inserts for the relevant table.
This patch change the default from INSERT (option 2 above) to COPY
(option 1) by:
Adding additional flag to the backup.sh script -c,
if this flag is set then the pg_dump is called with the --column-inserts
flag, otherwise not.
The default behaviour is to backup without --column-inserts in order to
generate COPY statements rather than INSERT statements per table row for
easy and efficient restore.
Change-Id: Iac6b8a6762868b9535f24cc9c496e11722485885
Signed-off-by: Eli Mesika <emesika(a)redhat.com>
---
M backend/manager/dbscripts/backup.sh
1 file changed, 11 insertions(+), 3 deletions(-)
Approvals:
Yair Zaslavsky: Verified; Looks good to me, approved
--
To view, visit http://gerrit.ovirt.org/8134
To unsubscribe, visit http://gerrit.ovirt.org/settings
Gerrit-MessageType: merged
Gerrit-Change-Id: Iac6b8a6762868b9535f24cc9c496e11722485885
Gerrit-PatchSet: 3
Gerrit-Project: ovirt-engine
Gerrit-Branch: master
Gerrit-Owner: Eli Mesika <emesika(a)redhat.com>
Gerrit-Reviewer: Allon Mureinik <amureini(a)redhat.com>
Gerrit-Reviewer: Eli Mesika <emesika(a)redhat.com>
Gerrit-Reviewer: Yair Zaslavsky <yzaslavs(a)redhat.com>