<div dir="ltr"><div><span style="font-size:12.8px">I'm using online_recovery </span><b style="font-size:12.8px">only</b><span style="font-size:12.8px"> for full copy purposes - if node gets corrupted. pcp_recovery_node will execute recovery_1st_stage command which using pg_basebackup (full copy).</span><br></div><div><span style="font-size:12.8px"><br></span></div><div><span style="font-size:12.8px">but for switchover between the nodes (<b>none</b> of the nodes gets corrupted) - just for <b>switch roles:</b></span></div><div><span style="font-size:12.8px"><b>1. shutdown the primary.</b></span></div><div><span style="font-size:12.8px"><b>2. pgpool promotes the secondary.</b></span></div><div><span style="font-size:12.8px"><b>3. perform pcp_attach_node (old primary) which calls the failback.sh:</b></span></div><div> the failback.sh does exactly what you describe:<div> a.<span style="font-size:12.8px">pg_start_backup()</span></div><div><span style="font-size:12.8px"> b. rsync (should be fast)</span></div><div><span style="font-size:12.8px"> c. pg_stop_</span><span style="font-size:12.8px">backup()</span></div></div><div><span style="font-size:12.8px"> d. creates recovery.conf</span></div><div><span style="font-size:12.8px"> e. start the node.</span></div><div><br></div><div><span style="font-size:12.8px"><br></span></div><div><span style="font-size:12.8px">any ideas or comments??</span></div><div><span style="font-size:12.8px"><br></span></div><div><span style="font-size:12.8px"><br></span></div><div><span style="font-size:12.8px">Thanks,</span></div><div><span style="font-size:12.8px">cohavisi</span></div><div><span style="font-size:12.8px"><br></span></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Tue, Mar 1, 2016 at 3:37 PM, Tatsuo Ishii <span dir="ltr"><<a href="mailto:ishii@postgresql.org" target="_blank">ishii@postgresql.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">> OK....Thanks!<br>
><br>
> I'm trying to implement an failover/failback on the nodes:<br>
> 1. primary node gets down.<br>
> 2. pgpool promotes the secondary node - make it primary.<br>
> 3. by attaching the failed node (old primary) - the failback.sh is called<br>
> and recovering the failed node (using rsync - much more faster) and make it<br>
> online secondary!<br>
<br>
</span>I don't know what failback.sh is doing but if it just runs rsync, it's<br>
not safe. You should use pg_start_backup()/pg_stop_backup().<br>
<br>
BTW, if rsync is much faster for you, why don't you use it for online<br>
recovery as well?<br>
<span class=""><br>
> from what you are saying...<br>
</span>> just to make sure, I *can not* use the failback.sh script (which called by<br>
<span class="">> pcp_attach_node) in order "recover" the node and make it online (as<br>
> scondary).<br>
<br>
</span>Ok, but failback.sh is not supposed to do what you want.<br>
I recommend you to look into follow master command.<br>
<div class="HOEnZb"><div class="h5"><br>
Best regards,<br>
--<br>
Tatsuo Ishii<br>
SRA OSS, Inc. Japan<br>
English: <a href="http://www.sraoss.co.jp/index_en.php" rel="noreferrer" target="_blank">http://www.sraoss.co.jp/index_en.php</a><br>
Japanese:<a href="http://www.sraoss.co.jp" rel="noreferrer" target="_blank">http://www.sraoss.co.jp</a><br>
<br>
> Thanks,<br>
> cohavisi<br>
><br>
> On Tue, Mar 1, 2016 at 2:15 PM, Tatsuo Ishii <<a href="mailto:ishii@postgresql.org">ishii@postgresql.org</a>> wrote:<br>
><br>
>> I'm not sure what you want to do (especialy I'm confused by<br>
>> "secondary": what does it mean?). Have you taken look at follow master<br>
>> script?<br>
>><br>
>> Anyway...<br>
>><br>
>> pcp_attach_node should be used for the case PostgreSQL server is<br>
>> online and ready to use. Not for recovering a PostgreSQL server.<br>
>><br>
>> Best regards,<br>
>> --<br>
>> Tatsuo Ishii<br>
>> SRA OSS, Inc. Japan<br>
>> English: <a href="http://www.sraoss.co.jp/index_en.php" rel="noreferrer" target="_blank">http://www.sraoss.co.jp/index_en.php</a><br>
>> Japanese:<a href="http://www.sraoss.co.jp" rel="noreferrer" target="_blank">http://www.sraoss.co.jp</a><br>
>><br>
>> > Hi,<br>
>> > Thanks for your replay...<br>
>> > I do use online recovery in case a full recovery is needed (using<br>
>> > pg_basebackup - via pcp_recovery_node).<br>
>> > but I added an ability to perform a switchover between the nodes using<br>
>> > stop/detach primary - failover occurs and reattach it as secondary (using<br>
>> > failback script).<br>
>> > but as the failback finished the pgpool does not attach it as secondary!!<br>
>> ><br>
>> ><br>
>> > Can you please advice?<br>
>> ><br>
>> > cohavisi<br>
>> ><br>
>> ><br>
>> > On Tue, Mar 1, 2016 at 10:41 AM, Tatsuo Ishii <<a href="mailto:ishii@postgresql.org">ishii@postgresql.org</a>><br>
>> wrote:<br>
>> ><br>
>> >> You should use online recovery instead of pcp_attach_node.<br>
>> >><br>
>> >> Best regards,<br>
>> >> --<br>
>> >> Tatsuo Ishii<br>
>> >> SRA OSS, Inc. Japan<br>
>> >> English: <a href="http://www.sraoss.co.jp/index_en.php" rel="noreferrer" target="_blank">http://www.sraoss.co.jp/index_en.php</a><br>
>> >> Japanese:<a href="http://www.sraoss.co.jp" rel="noreferrer" target="_blank">http://www.sraoss.co.jp</a><br>
>> >><br>
>> >> > Hi,<br>
>> >> > I have a Hugh problem regarding attaching a node (as secondary) to the<br>
>> >> pool<br>
>> >> > after I performing pcp_attach_node.<br>
>> >> ><br>
>> >> > after failover is being completed successfully and valid primary node<br>
>> is<br>
>> >> > active, i'm performing an *pcp_attach (via sql)* to the faulty node in<br>
>> >> > order to failback as secondary!<br>
>> >> ><br>
>> >> > *select pcp_attach_node (0,'10.10.61.99',1200,9898,'*****','*****') *<br>
>> >> ><br>
>> >> > during this command, a failback script is being executed and performs<br>
>> the<br>
>> >> > following:<br>
>> >> > 1. rsync between the DB nodes.<br>
>> >> > 2. create recovery.conf.<br>
>> >> > 3. startup the node(as secondary).<br>
>> >> ><br>
>> >> > *the failback could take for 20 min to finish.*<br>
>> >> ><br>
>> >> > after the failback finished *successfully* (exit status 0) and the<br>
>> node<br>
>> >> > started as *secondary* (according to postgres) - streaming<br>
>> replication.<br>
>> >> ><br>
>> >> > *the pgpool reportes the node status from 1 to 3 (instead of 2).*<br>
>> >> ><br>
>> >> > *** when failback finished early (less then few min) the pgpool<br>
>> reports<br>
>> >> the<br>
>> >> > node status as 2 - as aspected.*<br>
>> >> ><br>
>> >> ><br>
>> >> > please advice regarding this issue...<br>
>> >> ><br>
>> >> ><br>
>> >> > *Thanks,*<br>
>> >> > *cohavisi*<br>
>> >><br>
>><br>
</div></div></blockquote></div><br></div>