<html><head></head><body><div style="font-family:verdana, helvetica, sans-serif;font-size:13px;"><div></div>
<div>Thank you Scott as always!</div><div><br></div><div>Well, we want to "cluster" (load balancing, round robin) the Apache httpd RP (reverse proxy) nodes, since SP tights to the RP, it needs to be clustered with shared storageservice.</div><div><br></div><div>Regards</div><div>William<br></div><div><br></div>
<div id="ydpc9f11432yahoo_quoted_9900730556" class="ydpc9f11432yahoo_quoted">
<div style="font-family:'Helvetica Neue', Helvetica, Arial, sans-serif;font-size:13px;color:#26282a;">
<div>
On Tuesday, February 20, 2018, 2:54:39 PM EST, Cantor, Scott <cantor.2@osu.edu> wrote:
</div>
<div><br></div>
<div><br></div>
<div><div dir="ltr">> Besides RAC, we also tried memcached as StorageService, similar results:<br clear="none">> when one of memcached is offline, SP would just error out, doesn't seem to<br clear="none">> skip the bad one and try the other memcached.<br clear="none"><br clear="none">That would be entirely up to the memcache library.<br clear="none"><br clear="none">I already verified that the code doesn't care if it can't update the timestamp, so it's generally irrelevant for it to fail in that spot as long as it doesn't hang.<br clear="none"><br clear="none">It is also, almost invariably, pointless to cluster the SP.<div class="ydpc9f11432yqt1264614730" id="ydpc9f11432yqtfd65097"><br clear="none"><br clear="none">-- Scott<br clear="none"><br clear="none"></div></div></div>
</div>
</div></div></body></html>