[Bugs] [Bug 1323926] New: volume get functionality returns null value for few volume options
bugzilla at redhat.com
bugzilla at redhat.com
Tue Apr 5 06:02:04 UTC 2016
https://bugzilla.redhat.com/show_bug.cgi?id=1323926
Bug ID: 1323926
Summary: volume get functionality returns null value for few
volume options
Product: Red Hat Gluster Storage
Version: 3.1
Component: glusterfs-server
Sub Component: glusterd
Keywords: Triaged
Severity: medium
Assignee: amukherj at redhat.com
Reporter: sasundar at redhat.com
QA Contact: storage-qa-internal at redhat.com
CC: amukherj at redhat.com, bugs at gluster.org
Depends On: 1322211
+++ This bug was initially created as a clone of Bug #1322211 +++
Description of problem:
-----------------------
gluster volume get functionality returns a null value, even those options have
a default value which is non-null value
Version-Release number of selected component (if applicable):
-------------------------------------------------------------
gluster-3.7.9
How reproducible:
-----------------
Always
Steps to Reproduce:
-------------------
1. Get the default value of bind-insecure and allow-insecure
# gluster volume get <vol-name> bind-insecure
Actual results:
---------------
gluster volume get returns null value for few option eventhough they have
default options.
The default value of bind-insecure is off, but the volume get returns 'null'
value
Expected results:
-----------------
gluster volume get should get the default value for the options
Additional info:
----------------
I have filtered out the options for which 'gluster volume get' returns null
value. Few of the options the null value looks valid.
For eg, auth.reject & auth.allow - which haven't got any values and its
reasonable to display null values for those
But client.bind-insecure & server.allow-insecure has got a default values and
those should be printed rather than null value
[root at dhcp37-52 ~]# gluster volume get test all | grep \(null\)
cluster.subvols-per-directory (null)
cluster.rsync-hash-regex (null)
cluster.extra-hash-regex (null)
cluster.local-volume-name (null)
cluster.switch-pattern (null)
cluster.read-subvolume (null)
cluster.data-self-heal-algorithm (null)
cluster.quorum-count (null)
diagnostics.brick-logger (null)
diagnostics.client-logger (null)
diagnostics.brick-log-format (null)
diagnostics.client-log-format (null)
encryption.master-key (null)
network.tcp-window-size (null)
network.tcp-window-size (null)
auth.reject (null)
transport.keepalive (null)
server.allow-insecure (null)
features.grace-timeout (null)
server.ssl (null)
server.own-thread (null)
ssl.own-cert (null)
ssl.private-key (null)
ssl.ca-list (null)
ssl.crl-path (null)
ssl.certificate-depth (null)
ssl.cipher-list (null)
ssl.dh-param (null)
ssl.ec-curve (null)
features.limit-usage (null)
debug.exclude-ops (null)
debug.include-ops (null)
debug.error-failure (null)
debug.error-number (null)
debug.error-fops (null)
nfs.exports-auth-enable (null)
nfs.auth-refresh-interval-sec (null)
nfs.auth-cache-ttl-sec (null)
changelog.changelog-dir (null)
features.trash-eliminate-path (null)
locks.trace (null)
client.bind-insecure (null)
--- Additional comment from SATHEESARAN on 2016-03-29 23:04:37 EDT ---
[root@ ~]# gluster volume get test all | grep \(null\)
cluster.subvols-per-directory (null)
cluster.rsync-hash-regex (null)
cluster.extra-hash-regex (null)
cluster.local-volume-name (null)
cluster.switch-pattern (null)
cluster.read-subvolume (null)
cluster.data-self-heal-algorithm (null)
cluster.quorum-count (null)
diagnostics.brick-logger (null)
diagnostics.client-logger (null)
diagnostics.brick-log-format (null)
diagnostics.client-log-format (null)
encryption.master-key (null)
network.tcp-window-size (null)
network.tcp-window-size (null)
auth.reject (null)
transport.keepalive (null)
server.allow-insecure (null)
features.grace-timeout (null)
server.ssl (null)
server.own-thread (null)
ssl.own-cert (null)
ssl.private-key (null)
ssl.ca-list (null)
ssl.crl-path (null)
ssl.certificate-depth (null)
ssl.cipher-list (null)
ssl.dh-param (null)
ssl.ec-curve (null)
features.limit-usage (null)
debug.exclude-ops (null)
debug.include-ops (null)
debug.error-failure (null)
debug.error-number (null)
debug.error-fops (null)
nfs.exports-auth-enable (null)
nfs.auth-refresh-interval-sec (null)
nfs.auth-cache-ttl-sec (null)
changelog.changelog-dir (null)
features.trash-eliminate-path (null)
locks.trace (null)
client.bind-insecure (null)
--- Additional comment from SATHEESARAN on 2016-03-29 23:06:35 EDT ---
Ignore the volume options listing in comment0 and comment1, as the formatting
looks pretty bad and attaching the text file for the list of options
--- Additional comment from SATHEESARAN on 2016-03-29 23:07 EDT ---
--- Additional comment from SATHEESARAN on 2016-03-30 01:34:18 EDT ---
Apologies for incorrect statement about the default value of
client.bind-insecure in comment0
The default value of bind-insecure is 'on', yet 'gluster volume get' returns
the value as '(null)'
--- Additional comment from Atin Mukherjee on 2016-03-30 01:50:48 EDT ---
Logic of gluster volume get functionality is as follows:
1. Code first checks for if the key is configured in volinfo->dict which
indicates an explicit volume set command was triggered for this option.
2. If the key is not configured in 1, then it looks for the default value in
the volume map entry table maintained by glusterd.
3. If default value is not configured in VME table in glusterd, the code loads
the respective translator and checks for the default value.
In all the above three cases if the code doesn't find a match a null value is
returned.
The options what you have listed here falls into the same category as explained
in the above three points and glusterd doesn't find the default value for any
of them.
IMHO, the volume get logic still holds true. Until and unless the respective
options are configured with default value in either VME table or the
translators glusterd will continue to show null values for them.
--- Additional comment from SATHEESARAN on 2016-03-30 04:33:48 EDT ---
(In reply to Atin Mukherjee from comment #5)
> Logic of gluster volume get functionality is as follows:
>
> 1. Code first checks for if the key is configured in volinfo->dict which
> indicates an explicit volume set command was triggered for this option.
> 2. If the key is not configured in 1, then it looks for the default value in
> the volume map entry table maintained by glusterd.
> 3. If default value is not configured in VME table in glusterd, the code
> loads the respective translator and checks for the default value.
>
> In all the above three cases if the code doesn't find a match a null value
> is returned.
>
> The options what you have listed here falls into the same category as
> explained in the above three points and glusterd doesn't find the default
> value for any of them.
>
> IMHO, the volume get logic still holds true. Until and unless the respective
> options are configured with default value in either VME table or the
> translators glusterd will continue to show null values for them.
Thanks for the clean explanation.
Then it should be the corresponding translators should take care of the default
value for the options.
Referenced Bugs:
https://bugzilla.redhat.com/show_bug.cgi?id=1322211
[Bug 1322211] volume get functionality returns null value for few volume
options
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug https://bugzilla.redhat.com/token.cgi?t=h61m0TpTQV&a=cc_unsubscribe
More information about the Bugs
mailing list