mongodb single server
Samuel Carlier
i have a question about durability with a single server.
i have only 1 server to put mongodb on. what i read in this group is
that a min of 2 servers is recommend.
the question is what kind of dataloss is there with a server crash /
unclean termination
mongodb does flush data to disk every minute right? so when there is a
crash we have a chance to loss 1min of data?.
Michael Dirolf
1.8, but until then you probably shouldn't be running w/ just a single
node in production.
> --
> You received this message because you are subscribed to the Google Groups "mongodb-user" group.
> To post to this group, send email to mongod...@googlegroups.com.
> To unsubscribe from this group, send email to mongodb-user...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/mongodb-user?hl=en.
>
>
Andreas Jung
Hash: SHA1
http://blog.mongodb.org/post/381927266/what-about-durability
If you need shorter fsync intervals -> adjust it (some commandline
parameter - check the commandline docs).
- -aj
- --
ZOPYX Limited | zopyx group
Charlottenstr. 37/1 | The full-service network for Zope & Plone
D-72070 T�bingen | Produce & Publish
www.zopyx.com | www.produce-and-publish.com
- ------------------------------------------------------------------------
E-Publishing, Python, Zope & Plone development, Consulting
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/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=R04c
-----END PGP SIGNATURE-----
Markus Gattol
only loose data but much worse, your data set might become inconsistent.
When you are talking about at least two nodes, you are talking about a
master/slave setup. However, this has now been superseded by replica
sets where a minimum of three shared-nothing nodes (one primary and two
secondaries) is recommended (this has to do with automatic
failover/recovery and the voting procedure involved).
Michael Dirolf
> yup, I am not sure but I think Mike was indicating that you could not
> only loose data but much worse, your data set might become inconsistent.
I was indicating that you can lose more than a second of data.
> When you are talking about at least two nodes, you are talking about a
> master/slave setup. However, this has now been superseded by replica
> sets where a minimum of three shared-nothing nodes (one primary and two
> secondaries) is recommended (this has to do with automatic
> failover/recovery and the voting procedure involved).
You can deploy replica sets with only 2 physical nodes, the minimum is not 3.
Markus Gattol
>> yup, I am not sure but I think Mike was indicating that you could
>> not only loose data but much worse, your data set might become
>> inconsistent.
Michael> I was indicating that you can lose more than a second of data.
one second? You mean one minute no? (flush to disk happens every 60
seconds per default)
>> When you are talking about at least two nodes, you are talking
>> about a master/slave setup. However, this has now been superseded
>> by replica sets where a minimum of three shared-nothing nodes (one
>> primary and two secondaries) is recommended (this has to do with
>> automatic failover/recovery and the voting procedure involved).
Michael> You can deploy replica sets with only 2 physical nodes, the
Michael> minimum is not 3.
Sure. I said 3 is "recommended" otherwise an arbiter should be set up
http://www.markus-gattol.name/ws/mongodb.html#do_i_need_an_arbiter