Hello!

It's amazing to find most of your problems in Google, posted by other people 
years ago, but still no real solution; so let me try once more:

Simple question: How do I col-locate clones, specifically "interleaved=true" 
clones?

Remark: The documentation for "interleave" on the wiki is very hard (up to 
"impossible") to understand as people had complained about three years ago, but 
amazingly the description is still the same.

Why would I want it?: If you run OCFS filesystems on LVM, you naturally want to 
mount the filesystem where the LVs are present (and preferrably _after_ that).

While the ordering seems to work (leaving the popular "transitive ordering" 
issue aside), colocation seems to cause warnings like this:

Cannot pair prm_CFS:0 with instance of cln_cLVM

for

colocation col_OCFS_cVG inf: _rsc_set_ ( cln_CFS ) cln_cLVM

(The intention was to run the filesystem RA where the LVM RA had been 
successfully started (and thus provided a shared logical volume).

So maybe co-location for clones is implicit with the ordering constraint?:

order ord_cVG_CFS inf: cln_cLVM ( cln_CFS )

(That is for each node only start an instance of cln_CFS after an instance of 
cln_cLVM had been started successfully)

I still don't understand, and the construct that doesn't seem to be working can 
be found in the CTDB wiki at linux-ha.org.

So can anybody provide a sample configuration using the following stack of RAs?:

DLM (Distributed Lock Manager)
O2CB (OCFS2), needs DLM
cLVM needs DLM
LVM-LV needs cLVM
OCFS2-filesystem needs both, O2CB and LVM-LV

The pattern should be flexible enough to allow both, OFCS on top of an LV, as 
well as OCFS directly on a shared disk. And the pattern should only define 
contraints that are necessary, i.e. do not put everything in a group and clone 
that group.

While this seems to be a standard pattern of use, it's amazing how 
documentation works around configuring that (correctly).

I still believe that someone should be able to answer this question...

Regards,
Ulrich


_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to