Restart from....? (DRP)

Niels Dettenbach nd at syndicat.com
Mon Jun 18 05:19:27 EDT 2018


Am Montag, 18. Juni 2018, 10:48:16 CEST schrieb Albert Shih:
> Everything seem working fine, until I try to send the dataset on other
> server. I just cannot send a zfs snapshot from this server to another. If
> the dataset are small that's OK, but with the mailbox (~4To) the zfs
> command just hang after 10-40 minutes during 1-10 minutes, come back work
> during 1 or 2 hours and hang again etc.

Ahh,
yes,

we have local snapshots and a second ZFS machine for ZFS replication (incl. 
snapshots) which is running in the background - so the snapshots are done 
locally and send "in backround" over the network to another location. if just 
the machine but not the disks break, we can sue the local disk set within a 
new machine to start over. if the whole sites burn down, the disks (or 
temporarily by iSCSI, NFS or samba) could be used to start over in/with new 
hardware.

in smaller on-site setups we use i.e. FreeNAS as the "FreeBSD distribution" 
for easier management (even trough less skilled IT persons). this allows us 
to run jails with i.e. cyrus (capsulated and backed up too) which can be 
handled "by click" btw. this means the cyrus (jails) are running in ZFS too.

> Yes we using puppet, reinstalling the system and configuration are easy.
> The hard part are the data. 
this depends from the storage (on network like NAS or SAN or "locally"). by 
principle:

 - mounting or copying the pool (usually the largest part)
 - reimport database

i.e. similiar to:
	https://forum.open-xchange.com/showthread.php?3512-Simple-Cyrus-mailbox-migration

or

	http://www.monoplan.de/cyrus_imap_migration.html (german)

 - reconstruct -f  (i use "just" reconstruct -f as this runs the whole pool 
too)

then your cyrus should be fine again. the reconstruct -f (forced) mail reads 
the pool data (folder by folder, mail by mail) and "fixes" any inconsistencies 
with the "database" (due to the "hot" state of the backup - not shutted down 
for backup). the cyrus databases seems quite robust to that (compared to most 
other database systems).

> I'm a bit new with cyrus so...  All I can say is the replication seem to
> works well. I got 
thanks for this info. will try cyrus replication soon for testing purposes. 
ß)

> I'll will try today to see if it's easy or not to restart with a slave by
> cloning it. 
i'm new to replication, but afaik it should be easy to make a slave to a new 
master just by reconfiguration /cyrus.conf, some imapd.conf flags)  i assume.



hth a bit,
good luck,


niels.




-- 
 ---
 Niels Dettenbach
 Syndicat IT & Internet
 http://www.syndicat.com
 PGP: https://syndicat.com/pub_key.asc
 ---
 





More information about the Info-cyrus mailing list