<div dir="ltr"><div class="gmail_default" style="font-family:arial,helvetica,sans-serif"><br></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Wed, May 7, 2014 at 6:10 PM, James Le Cuirot <span dir="ltr"><<a href="mailto:chewi@aura-online.co.uk" target="_blank">chewi@aura-online.co.uk</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">Hello,<br>
<br>
I know this has been asked before but I felt that it wasn't fully<br>
answered and I think the situation may have changed in 3.5.<br>
<br>
I have set up geo-replication between two machines on my LAN for<br>
testing. Both are using NTP and the clocks are definitely in sync.<br>
CRAWL STATUS reports Changelog Crawl. When I make a change on the<br>
master, it takes up to a minute (sometimes less) for the slave to<br>
notice.<br></blockquote><div><br></div><div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">That's because the change detection interval is 60 seconds by default[1] in 3.5. This has been changed to 15 seconds lately.</div>
</div><div><br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
<br>
Now I understand that geo-replication will always have some delay, not<br>
least because it is asynchronous, but given these are tiny changes with<br>
practically no other activity going on, I was expecting it to be a<br>
little more responsive. Even in production, there will very little<br>
traffic so some additional resource usage to speed things up would not<br>
be an issue. Is this configurable at all?<br></blockquote><div><br></div><div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">It's configurable. Try this:</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">
<br></div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif"># gluster volume set <volume> rollover-time 1</div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif"><br></div>
<div class="gmail_default" style="font-family:arial,helvetica,sans-serif">to indentify changes (followed by syncing it to the slave) every second. This will definitely speed up replication.</div><br></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
<br>
I've seen it explained previously that inotify does not scale well and<br>
gluster has taken a more efficient approach. I hadn't expected this to<br>
come at the cost of such long delays though. We're only planning to<br>
have two nodes, a master and a slave, and I've also seen it said that<br>
gluster provides little benefit over a simple periodic call to rsync<br>
under such setups. Combine that with inotify and the rsync solution<br>
even starts to look favourable.<br>
<br>
Lowering this delay is not a deal-breaker for us, it's just that it<br>
seems unnecessarily long. I'd appreciate hearing your thoughts.<br></blockquote><div><br></div><div><div class="gmail_default" style="font-family:arial,helvetica,sans-serif">[1]: <a href="https://github.com/gluster/glusterfs/blob/release-3.5/xlators/features/changelog/src/changelog.c#L1556">https://github.com/gluster/glusterfs/blob/release-3.5/xlators/features/changelog/src/changelog.c#L1556</a></div>
</div></div></div></div>