On 23 Oct 2019, at 04:00, I M <isabella.ghiurea(a)nrc-cnrc.gc.ca>
wrote:
Mark, SImon thank you for feedback!
The connections between master to master is good same location but the one from to new
DS slave on different location is poor, so this should imply to have 2agreements , also we
are using memberof pluging , for each master host the memberof is excluded from
replication ( cfg as fractional replication)
and for master to slave ( I assume member off will not be excluded will be full
replication)
Than would make sense to have only one master to slave full replication agreement?
Think of it a different way - I like pictures with this
[ Write A ] <<-- repl -->> [ Write B ]
\
\ link a
\
\----> [ RO C ]
In the first scenario if something happens to Write A, or link a, or the repl link you now
isolate RO C such as:
[ Write A ] [ Write B ]
\
\ link a
\
\----> [ RO C ]
[ Write A ] <<-- repl -->> [ Write B ]
\
[ RO C ]
This way changes to Write B or Write A won't make it to C. And if the link is
"unreliable" then this is more likely to happen.
But if you have multipath:
[ Write A ] <<-- repl -->> [ Write B ]
\ /
\ link a link b /
\ /
\----> [ RO C ] <---/
Issues on repl, link a or link b will shield you from issues because the other path will
be selected to C.
So I'd advise you always try to have at least 2 incoming replication agreements to any
server in any topology
_______________________________________________
389-users mailing list -- 389-users(a)lists.fedoraproject.org
To unsubscribe send an email to 389-users-leave(a)lists.fedoraproject.org
Fedora Code of Conduct:
https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines:
https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives:
https://lists.fedoraproject.org/archives/list/389-users@lists.fedoraproje...
—
Sincerely,
William Brown
Senior Software Engineer, 389 Directory Server
SUSE Labs