Hi Jeff,

I need a little clarification, if I get the things right. You wrote that both trailing and leading commas are removed by the script. The script checks if there is comma is in the beginning or at the end and if yes script removes it, is that right?

I can only reproduce trailing comma. How can you reproduce leading comma?

Thanks for your answer in advance.

I cannot dup this issue any longer but I believe that you get the leading comma when your volume name is "gv0" (due to a bug in ambari where gv0 is hard-coded as this prop value.)

Ok, just to be 100% clear:

I tested script in rhs-hadoop-install 2.28,2.29 and 2.39. 

The script 2.28 add a trailing comma, if name of the volume is "gv0".
The script 2.29 add no comma, but if the comma is already there, it does not remove it.
The script in 2.39 does not create a trailing comma and removes a trailing comma if it is already there in core-site.xml

My question is: Should the script also work like the above with leading comma? (it does not matter that I can't reproduce it, I can add the comma there manually).

Thank you

Jakub,  Thanks for your extra testing!  If the user manually edits the fs.glusterfs.volumes property with leading and/or trailing commas, then all leading and all trailing commas are deleted when bin/ambari_config_update.sh removes the target volume name from the list of volume names.

Verified on the last version (2.44). No leading and/or trailing commas are present in core-site.xml. Moving to VERIFIED

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://rhn.redhat.com/errata/RHEA-2015-0761.html