So I suppose ctime is enough to consider it whether a good iatt or not.
And why we also include ia_nlink in function gf_zero_fill_stat and
gf_is_zero_filled_stat ?
From my investigation, if set ia_nlink to 0, if kernel read the attr with flag
RCU, kernel will check the ia_nlink field, when do LINK operation, it will lead
to error of “files is not exist”.
if (inode->i_nlink == 0 && !(inode->i_state & I_LINKABLE))
error = -ENOENT;
Best Regards,
George
From: [email protected]
[mailto:[email protected]] On Behalf Of Pranith Kumar Karampuri
Sent: Wednesday, January 24, 2018 4:15 PM
To: Lian, George (NSB - CN/Hangzhou) <[email protected]>
Cc: Zhou, Cynthia (NSB - CN/Hangzhou) <[email protected]>;
[email protected]; Li, Deqian (NSB - CN/Hangzhou)
<[email protected]>; Sun, Ping (NSB - CN/Hangzhou)
<[email protected]>
Subject: Re: [Gluster-devel] a link issue maybe introduced in a bug fix " Don't
let NFS cache stat after writes"
If ctime is zero, no xlator should consider it as a good iatt. The fact that
this is happening means some xlator is not doing proper checks in code. We need
to find what that xlator is and fix it. Internet in our new office is not
working so I'm not able to have call today with you guys. What I would do is to
put logs in lookup, link, fstat, stat calls to see if anyone unwound iatt with
ia_nlink count as zero but ctime as nonzero.
On 24 Jan 2018 1:03 pm, "Lian, George (NSB - CN/Hangzhou)"
<[email protected]<mailto:[email protected]>> wrote:
Hi, Pranith Kumar,
Can you tell me while need set buf->ia_nlink to “0”in function
gf_zero_fill_stat(), which API or Application will care it?
If I remove this line and also update corresponding in function
gf_is_zero_filled_stat,
The issue seems gone, but I can’t confirm will lead to other issues.
So could you please double check it and give your comments?
My change is as the below:
gf_boolean_t
gf_is_zero_filled_stat (struct iatt *buf)
{
if (!buf)
return 1;
/* Do not use st_dev because it is transformed to store the xlator id
* in place of the device number. Do not use st_ino because by this time
* we've already mapped the root ino to 1 so it is not guaranteed to be
* 0.
*/
// if ((buf->ia_nlink == 0) && (buf->ia_ctime == 0))
if (buf->ia_ctime == )
return 1;
return 0;
}
void
gf_zero_fill_stat (struct iatt *buf)
{
// buf->ia_nlink = 0;
buf->ia_ctime = 0;
}
Thanks & Best Regards
George
From: Lian, George (NSB - CN/Hangzhou)
Sent: Friday, January 19, 2018 10:03 AM
To: Pranith Kumar Karampuri <[email protected]<mailto:[email protected]>>;
Zhou, Cynthia (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>
Cc: Li, Deqian (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>;
[email protected]<mailto:[email protected]>; Sun, Ping (NSB -
CN/Hangzhou) <[email protected]<mailto:[email protected]>>
Subject: RE: [Gluster-devel] a link issue maybe introduced in a bug fix " Don't
let NFS cache stat after writes"
Hi,
>>> Cool, this works for me too. Send me a mail off-list once you are available
>>> and we can figure out a way to get into a call and work on this.
Have you reproduced the issue per the step I listed in
https://bugzilla.redhat.com/show_bug.cgi?id=1531457 and last mail?
If not, I would like you could try it yourself , which the difference between
yours and mine is just create only 2 bricks instead of 6 bricks.
And Cynthia could have a session with you if you needed when I am not available
in next Monday and Tuesday.
Thanks & Best Regards,
George
From:
[email protected]<mailto:[email protected]>
[mailto:[email protected]] On Behalf Of Pranith Kumar Karampuri
Sent: Thursday, January 18, 2018 6:03 PM
To: Lian, George (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>
Cc: Zhou, Cynthia (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>; Li, Deqian
(NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>;
[email protected]<mailto:[email protected]>; Sun, Ping (NSB -
CN/Hangzhou) <[email protected]<mailto:[email protected]>>
Subject: Re: [Gluster-devel] a link issue maybe introduced in a bug fix " Don't
let NFS cache stat after writes"
On Thu, Jan 18, 2018 at 12:17 PM, Lian, George (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>> wrote:
Hi,
>>>I actually tried it with replica-2 and replica-3 and then distributed
>>>replica-2 before replying to the earlier mail. We can have a debugging
>>>session if you are okay with it.
It is fine if you can’t reproduce the issue in your ENV.
And I has attached the detail reproduce log in the Bugzilla FYI
But I am sorry I maybe OOO at Monday and Tuesday next week, so debug session
will be fine to me at next Wednesday.
Cool, this works for me too. Send me a mail off-list once you are available and
we can figure out a way to get into a call and work on this.
Paste the detail reproduce log FYI here:
root@ubuntu:~# gluster peer probe ubuntu
peer probe: success. Probe on localhost not needed
root@ubuntu:~# gluster v create test replica 2 ubuntu:/home/gfs/b1
ubuntu:/home/gfs/b2 force
volume create: test: success: please start the volume to access data
root@ubuntu:~# gluster v start test
volume start: test: success
root@ubuntu:~# gluster v info test
Volume Name: test
Type: Replicate
Volume ID: fef5fca3-81d9-46d3-8847-74cde6f701a5
Status: Started
Snapshot Count: 0
Number of Bricks: 1 x 2 = 2
Transport-type: tcp
Bricks:
Brick1: ubuntu:/home/gfs/b1
Brick2: ubuntu:/home/gfs/b2
Options Reconfigured:
transport.address-family: inet
nfs.disable: on
performance.client-io-threads: off
root@ubuntu:~# gluster v status
Status of volume: test
Gluster process TCP Port RDMA Port Online Pid
------------------------------------------------------------------------------
Brick ubuntu:/home/gfs/b1 49152 0 Y 7798
Brick ubuntu:/home/gfs/b2 49153 0 Y 7818
Self-heal Daemon on localhost N/A N/A Y 7839
Task Status of Volume test
------------------------------------------------------------------------------
There are no active volume tasks
root@ubuntu:~# gluster v set test cluster.consistent-metadata on
volume set: success
root@ubuntu:~# ls /mnt/test
ls: cannot access '/mnt/test': No such file or directory
root@ubuntu:~# mkdir -p /mnt/test
root@ubuntu:~# mount -t glusterfs ubuntu:/test /mnt/test
root@ubuntu:~# cd /mnt/test
root@ubuntu:/mnt/test# echo "abc">aaa
root@ubuntu:/mnt/test# cp aaa bbb;link bbb ccc
root@ubuntu:/mnt/test# kill -9 7818
root@ubuntu:/mnt/test# cp aaa ddd;link ddd eee
link: cannot create link 'eee' to 'ddd': No such file or directory
Best Regards,
George
From:
[email protected]<mailto:[email protected]>
[mailto:[email protected]<mailto:[email protected]>]
On Behalf Of Pranith Kumar Karampuri
Sent: Thursday, January 18, 2018 2:40 PM
To: Lian, George (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>
Cc: Zhou, Cynthia (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>;
[email protected]<mailto:[email protected]>; Li, Deqian (NSB -
CN/Hangzhou) <[email protected]<mailto:[email protected]>>;
Sun, Ping (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>
Subject: Re: [Gluster-devel] a link issue maybe introduced in a bug fix " Don't
let NFS cache stat after writes"
On Thu, Jan 18, 2018 at 6:33 AM, Lian, George (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>> wrote:
Hi,
I suppose the brick numbers in your testing is six, and you just shut down the
3 process.
When I reproduce the issue, I only create a replicate volume with 2 bricks,
only let ONE brick working and set cluster.consistent-metadata on,
With this 2 test condition, the issue could 100% reproducible.
Hi,
I actually tried it with replica-2 and replica-3 and then distributed
replica-2 before replying to the earlier mail. We can have a debugging session
if you are okay with it.
I am in the middle of a customer issue myself(That is the reason for this delay
:-( ) and thinking of wrapping it up early next week. Would that be fine with
you?
16:44:28 :) ⚡ gluster v status
Status of volume: r2
Gluster process TCP Port RDMA Port Online Pid
------------------------------------------------------------------------------
Brick localhost.localdomain:/home/gfs/r2_0 49152 0 Y 5309
Brick localhost.localdomain:/home/gfs/r2_1 49154 0 Y 5330
Brick localhost.localdomain:/home/gfs/r2_2 49156 0 Y 5351
Brick localhost.localdomain:/home/gfs/r2_3 49158 0 Y 5372
Brick localhost.localdomain:/home/gfs/r2_4 49159 0 Y 5393
Brick localhost.localdomain:/home/gfs/r2_5 49160 0 Y 5414
Self-heal Daemon on localhost N/A N/A Y 5436
Task Status of Volume r2
------------------------------------------------------------------------------
There are no active volume tasks
root@dhcp35-190 - ~
16:44:38 :) ⚡ kill -9 5309 5351 5393
Best Regards,
George
From:
[email protected]<mailto:[email protected]>
[mailto:[email protected]<mailto:[email protected]>]
On Behalf Of Pranith Kumar Karampuri
Sent: Wednesday, January 17, 2018 7:27 PM
To: Lian, George (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>
Cc: Li, Deqian (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>;
[email protected]<mailto:[email protected]>; Zhou, Cynthia (NSB
- CN/Hangzhou)
<[email protected]<mailto:[email protected]>>; Sun, Ping
(NSB - CN/Hangzhou) <[email protected]<mailto:[email protected]>>
Subject: Re: [Gluster-devel] a link issue maybe introduced in a bug fix " Don't
let NFS cache stat after writes"
On Mon, Jan 15, 2018 at 1:55 PM, Pranith Kumar Karampuri
<[email protected]<mailto:[email protected]>> wrote:
On Mon, Jan 15, 2018 at 8:46 AM, Lian, George (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>> wrote:
Hi,
Have you reproduced this issue? If yes, could you please confirm whether it is
an issue or not?
Hi,
I tried recreating this on my laptop and on both master and 3.12 and I
am not able to recreate the issue :-(.
Here is the execution log:
https://paste.fedoraproject.org/paste/-csXUKrwsbrZAVW1KzggQQ
Since I was doing this on my laptop, I changed shutting down of the replica to
killing the brick process to simulate this test.
Let me know if I missed something.
Sorry, I am held up with some issue at work, so I think I will get some time
day after tomorrow to look at this. In the mean time I am adding more people
who know about afr to see if they get a chance to work on this before me.
And if it is an issue, do you have any solution for this issue?
Thanks & Best Regards,
George
From: Lian, George (NSB - CN/Hangzhou)
Sent: Thursday, January 11, 2018 2:01 PM
To: Pranith Kumar Karampuri <[email protected]<mailto:[email protected]>>
Cc: Zhou, Cynthia (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>;
[email protected]<mailto:[email protected]>; Li, Deqian (NSB -
CN/Hangzhou) <[email protected]<mailto:[email protected]>>;
Sun, Ping (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>
Subject: RE: [Gluster-devel] a link issue maybe introduced in a bug fix " Don't
let NFS cache stat after writes"
Hi,
Please see detail test step on
https://bugzilla.redhat.com/show_bug.cgi?id=1531457
How reproducible:
Steps to Reproduce:
1.create a volume name "test" with replicated
2.set volume option cluster.consistent-metadata with on:
gluster v set test cluster.consistent-metadata on
3. mount volume test on client on /mnt/test
4. create a file aaa size more than 1 byte
echo "1234567890" >/mnt/test/aaa
5. shutdown a replicat node, let's say sn-1, only let sn-0 worked
6. cp /mnt/test/aaa /mnt/test/bbb; link /mnt/test/bbb /mnt/test/ccc
BRs
George
From:
[email protected]<mailto:[email protected]>
[mailto:[email protected]] On Behalf Of Pranith Kumar Karampuri
Sent: Thursday, January 11, 2018 12:39 PM
To: Lian, George (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>
Cc: Zhou, Cynthia (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>;
[email protected]<mailto:[email protected]>; Li, Deqian (NSB -
CN/Hangzhou) <[email protected]<mailto:[email protected]>>;
Sun, Ping (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>
Subject: Re: [Gluster-devel] a link issue maybe introduced in a bug fix " Don't
let NFS cache stat after writes"
On Thu, Jan 11, 2018 at 6:35 AM, Lian, George (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>> wrote:
Hi,
>>> In which protocol are you seeing this issue? Fuse/NFS/SMB?
It is fuse, within mountpoint by “mount -t glusterfs …“ command.
Could you let me know the test you did so that I can try to re-create and see
what exactly is going on?
Configuration of the volume and the steps to re-create the issue you are seeing
would be helpful in debugging the issue further.
Thanks & Best Regards,
George
From:
[email protected]<mailto:[email protected]>
[mailto:[email protected]<mailto:[email protected]>]
On Behalf Of Pranith Kumar Karampuri
Sent: Wednesday, January 10, 2018 8:08 PM
To: Lian, George (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>
Cc: Zhou, Cynthia (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>; Zhong, Hua
(NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>>; Li, Deqian (NSB
- CN/Hangzhou) <[email protected]<mailto:[email protected]>>;
[email protected]<mailto:[email protected]>; Sun, Ping (NSB -
CN/Hangzhou) <[email protected]<mailto:[email protected]>>
Subject: Re: [Gluster-devel] a link issue maybe introduced in a bug fix " Don't
let NFS cache stat after writes"
On Wed, Jan 10, 2018 at 11:09 AM, Lian, George (NSB - CN/Hangzhou)
<[email protected]<mailto:[email protected]>> wrote:
Hi, Pranith Kumar,
I has create a bug on Bugzilla
https://bugzilla.redhat.com/show_bug.cgi?id=1531457
After my investigation for this link issue, I suppose your changes on
afr-dir-write.c with issue " Don't let NFS cache stat after writes" , your fix
is like:
--------------------------------------
if (afr_txn_nothing_failed (frame, this)) {
/*if it did pre-op, it will do post-op changing ctime*/
if (priv->consistent_metadata &&
afr_needs_changelog_update (local))
afr_zero_fill_stat (local);
local->transaction.unwind (frame, this);
}
In the above fix, it set the ia_nlink to ‘0’ if option consistent-metadata is
set to “on”.
And hard link a file with which just created will lead to an error, and the
error is caused in kernel function “vfs_link”:
if (inode->i_nlink == 0 && !(inode->i_state & I_LINKABLE))
error = -ENOENT;
could you please have a check and give some comments here?
When stat is "zero filled", understanding is that the higher layer protocol
doesn't send stat value to the kernel and a separate lookup is sent by the
kernel to get the latest stat value. In which protocol are you seeing this
issue? Fuse/NFS/SMB?
Thanks & Best Regards,
George
--
Pranith
--
Pranith
--
Pranith
--
Pranith
--
Pranith
--
Pranith
_______________________________________________
Gluster-devel mailing list
[email protected]
http://lists.gluster.org/mailman/listinfo/gluster-devel
