Hi,
Running Bucardo 5.1.1, I understand syncs now can get 'stalled' so a
broken sync no longer blocks the other syncs from being updated. Doing
some experiments I noticed the following:
When I shut down PostgreSQL on one of the nodes in a sync, the KID exits
and the MCP marks the affected db and sync as stalled. New KIDs get
started, and exit as long as PostgreSQL is down. Running 'bucardo
status' says the state of the sync is bad.
Then I start PostgreSQL again. The KID finished the syncrun and 'bucardo
status' says the state of the sync is good.
Bucardo keeps on running, but nothing happens anymore because the sync
is still stalled. 'bucardo status' reports the state of the sync as
good. Doing a restart marks the sync and db again as active and
everything returns to normal.
Is restarting the only way to 'unstall' a sync? Is the best/only way to
monitor if a sync is stalled doing a 'select status from sync'?
Kind regards
--
Hans van der Riet
_______________________________________________
Bucardo-general mailing list
[email protected]
https://mail.endcrypt.com/mailman/listinfo/bucardo-general
