Bug 1304465 - dnscache in libglusterfs returns 127.0.0.1 for 1st non-localhost request
Summary: dnscache in libglusterfs returns 127.0.0.1 for 1st non-localhost request
Keywords:
Status: CLOSED WORKSFORME
Alias: None
Product: GlusterFS
Classification: Community
Component: core
Version: mainline
Hardware: Unspecified
OS: Unspecified
medium
low
Target Milestone: ---
Assignee: Rinku
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2016-02-03 17:16 UTC by Niels de Vos
Modified: 2019-06-17 05:49 UTC (History)
4 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2019-06-17 05:49:02 UTC
Regression: ---
Mount Type: ---
Documentation: ---
CRM:
Verified Versions:
Embargoed:


Attachments (Terms of Use)

Description Niels de Vos 2016-02-03 17:16:01 UTC
Description of problem:
When 

Version-Release number of selected component (if applicable):


How reproducible:
100%

Steps to Reproduce:
1. setup a Gluster volume (server is called storage.example.com below)
2. on a non-Gluster server (client only):
   # mount -t glusterfs -o backup-volfile-servers=storage.example.com \
            localhost:/volume /mnt

Actual results:
Mounting will fail.

Expected results:
Mounting succeeds.

Additional info:
Enabling DEBUG logging with the additional "log-level=DEBUG" mount option shows that the 1st DNS lookup in common-utils.c:333:gf_resolve_ip6 returns 127.0.0.1 for the storage.example.com hostname.

Comment 1 Amar Tumballi 2019-05-09 10:09:05 UTC
Can we see if this is the behavior with glusterfs-6.x and take appropriate action? 

I have not heard about this issue in recent time, but would validate it before closing this.

Comment 2 Rinku 2019-06-13 07:45:54 UTC
Glusterfs version used for testing : v6.3

Servers used for testing : 

    Hostname : GlusterNode1.com, Ip :  192.168.1.10
    Hostname : GlusterNode2.com, Ip : 192.168.1.20

Following distributed volume was created : 

    # gluster v info devvol
    Volume Name: devvol
    Type: Distribute
    Volume ID: 2b3a4bb0-f8c2-40e5-b5f5-61b9e89ace45
    Status: Started
    Snapshot Count: 0
    Number of Bricks: 2
    Transport-type: tcp
    Bricks:
    Brick1: 192.168.1.10:/testxfs/brick-a1/brick
    Brick2: 192.168.1.20:/testxfs/brick-b1/brick
    Options Reconfigured:
    transport.address-family: inet
    nfs.disable: on

Fired the following command from the client : 

    # mount -t glusterfs -o backup-volfile-servers=GlusterNode2.com,log-level=DEBUG GlusterNode1.com:/devvol /mnt1


Result :  

    Was successfully able to mount the volume. 

Logs : 

    [2019-06-13 07:25:32.159993] D [MSGID: 0] [common-utils.c:532:gf_resolve_ip6] 0-resolver: returning ip-192.168.1.10 (port-24007) for hostname: GlusterNode1.com and port: 24007
    .
    .
    .
    [2019-06-13 07:25:36.391372] D [MSGID: 0] [common-utils.c:532:gf_resolve_ip6] 0-resolver: returning ip-192.168.1.20 (port-24007) for hostname: GlusterNode2.com and port: 24007

Comment 3 Rinku 2019-06-17 05:49:02 UTC
As this is no more reproducible on latest version (v6.3) we are closing this.


Note You need to log in before you can comment on or make changes to this bug.