<div dir="ltr">Can this get into 'FAQ' document somewhere? This is one of the major question asked all the time.<div><br></div><div>Regards,</div><div>Amar</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Jan 10, 2018 at 10:56 AM, Raghavendra Gowdappa <span dir="ltr"><<a href="mailto:rgowdapp@redhat.com" target="_blank">rgowdapp@redhat.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Sorry about the delayed response. Had to dig into the history to answer various "why"s.<br>
<span class=""><br>
----- Original Message -----<br>
> From: "Omar Kohl" <<a href="mailto:omar.kohl@iternity.com">omar.kohl@iternity.com</a>><br>
> To: <a href="mailto:gluster-users@gluster.org">gluster-users@gluster.org</a><br>
> Sent: Tuesday, December 26, 2017 6:41:48 PM<br>
> Subject: [Gluster-users] Exact purpose of network.ping-timeout<br>
><br>
> Hi,<br>
><br>
> I have a question regarding the "ping-timeout" option. I have been<br>
> researching its purpose for a few days and it is not completely clear to me.<br>
> Especially that it is apparently strongly encouraged by the Gluster<br>
> community not to change or at least decrease this value!<br>
><br>
> Assuming that I set ping-timeout to 10 seconds (instead of the default 42)<br>
> this would mean that if I have a network outage of 11 seconds then Gluster<br>
> internally would have to re-allocate some resources that it freed after the<br>
> 10 seconds, correct? But apart from that there are no negative implications,<br>
> are there? For instance if I'm copying files during the network outage then<br>
> those files will continue copying after those 11 seconds.<br>
><br>
> This means that the only purpose of ping-timeout is to save those extra<br>
> resources that are used by "short" network outages. Is that correct?<br>
<br>
</span>Basic purpose of ping-timer/heartbeat is to identify an unresponsive brick. Unresponsiveness can be caused due to various reasons like:<br>
* A deadlocked server. We no longer see too many instances of deadlocked bricks/server<br>
* Slow execution of fops in brick stack. For eg.,<br>
- due to lock contention. There have been some efforts to fix the lock contention on brick stack.<br>
- bad backend OS/filesystem. Posix health checker was an effort to fix this.<br>
- Not enough threads for execution etc<br>
Note that ideally its not the job of ping framework to identify this scenario and following the same thought process we've shielded the processing of ping requests on bricks from the costs of execution of requests to Glusterfs Program.<br>
<br>
* Ungraceful shutdown of network connections. For eg.,<br>
- hard shutdown of machine/container/VM running the brick<br>
- physically pulling out the network cable<br>
Basically all those different scenarios where TCP/IP doesn't get a chance to inform the other end that it is going down. Note that some of the scenarios of ungraceful network shutdown can be identified using TCP_KEEPALIVE and TCP_USERTIMEOUT [1]. However, at the time when heartbeat mechanism was introduced in Glusterfs, TCP_KEEPALIVE couldn't identify all the ungraceful network shutdown scenarios and TCP_USER_TIMEOUT was yet to be implemented in Linux kernel. One scenario which TCP_KEEPALIVE could identify was the exact scenario TCP_USER_TIMEOUT aims to solve - identifying an hard network shutdown when data is in transit. However there might be other limitations in TCP_KEEPALIVE which we need to test out before retiring heart beat mechanism in favor of TCP_KEEPALIVE and TCP_USER_TIMEOUT.<br>
<br>
The next interesting question would be why we need to identify an unresponsive brick. Various reasons why we need to do that would be:<br>
* To replace/fix any problems the brick might have<br>
* Almost all of the cluster translators - DHT, AFR, EC - wait for a response from all of their children - either successful or failure - before sending the response back to application. This means one or more slow/unresponsive brick can increase the latencies of fops/syscalls even though other bricks are responsive and healthy. However there are ongoing efforts to minimize the effect of few slow/unresponsive bricks [2]. I think principles of [2] can applied to DHT and AFR too.<br>
<br>
Some recent discussions on the necessity of ping framework in glusterfs can be found at [3].<br>
<br>
Having given all the above reasons for the existence of ping framework, its also important that ping-framework shouldn't bring down an otherwise healthy connection (False positives). Reasons are:<br>
* As pointed out by Joe Julian in another mail on this thread, each connection carries some state on bricks like locks/open-fds which is cleaned up on a disconnect. So, disconnects (even those followed by quick reconnects) are not completely transient to application. Though presence of HA layers like EC/AFR mitigates this problem to some extent, we still don't have a lock healing implementation in place. So, once Quorum number of AFR/EC children go down (though may not be all at once), locks are no longer held on bricks.<br>
* All the fops that are in transit in the time window starting from the time of disconnect till a successful reconnect are failed by rpc/transport layer. So, based on the configuration of volumes (whether AFR/EC/DHT prevent these errors from being seen by application), this *may* result in application seeing the error.<br>
<br>
IOW, disconnects are not lightweight and we need to avoid them whenever possible. Since the action on ping-timer expiry is to disconnect the connection, we suggest not have very low values to avoid spurious disconnections.<br>
<br>
[1] <a href="http://man7.org/linux/man-pages/man7/tcp.7.html" rel="noreferrer" target="_blank">http://man7.org/linux/man-<wbr>pages/man7/tcp.7.html</a><br>
[2] <a href="https://github.com/gluster/glusterfs/issues/366" rel="noreferrer" target="_blank">https://github.com/gluster/<wbr>glusterfs/issues/366</a><br>
[3] <a href="http://lists.gluster.org/pipermail/gluster-devel/2017-January/051938.html" rel="noreferrer" target="_blank">http://lists.gluster.org/<wbr>pipermail/gluster-devel/2017-<wbr>January/051938.html</a><br>
<div class="HOEnZb"><div class="h5"><br>
><br>
> If I am confident that my network will not have many 11 second outages and if<br>
> they do occur I am willing to incur those extra costs due to resource<br>
> allocation is there any reason not to set ping-timeout to 10 seconds?<br>
><br>
> The problem I have with a long ping-timeout is that the Windows Samba Client<br>
> disconnects after 25 seconds. So if one of the nodes of a Gluster cluster<br>
> shuts down ungracefully then the Samba Client disconnects and the file that<br>
> was being copied is incomplete on the server. These "costs" seem to be much<br>
> higher than the potential costs of those Gluster resource re-allocations.<br>
> But it is hard to estimate because there is not clear documentation what<br>
> exactly those Gluster costs are.<br>
><br>
> In general I would be very interested in a comprehensive explanation of<br>
> ping-timeout and the up- and downsides of setting high or low values for it.<br>
><br>
> Kinds regards,<br>
> Omar<br>
> ______________________________<wbr>_________________<br>
> Gluster-users mailing list<br>
> <a href="mailto:Gluster-users@gluster.org">Gluster-users@gluster.org</a><br>
> <a href="http://lists.gluster.org/mailman/listinfo/gluster-users" rel="noreferrer" target="_blank">http://lists.gluster.org/<wbr>mailman/listinfo/gluster-users</a><br>
><br>
______________________________<wbr>_________________<br>
Gluster-users mailing list<br>
<a href="mailto:Gluster-users@gluster.org">Gluster-users@gluster.org</a><br>
<a href="http://lists.gluster.org/mailman/listinfo/gluster-users" rel="noreferrer" target="_blank">http://lists.gluster.org/<wbr>mailman/listinfo/gluster-users</a><br>
</div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div>Amar Tumballi (amarts)<br></div></div></div></div></div>
</div>