I am resharing this, because as best I can tell, the support group didn't get it. (It may be that I got it because I manually un-archived Paul?)
♲ Paul Taylor
I've been offline in ~f terms since 16 September when the box the site was served from went bang (literally). A horrible sequence of events, which I won't go into, has led to the restoration of the site being a painfully slow and a slowly painful process. I am nearly there but am encountering some problems which I thought I'd ask about.
The vserver I am using now is a single-core with only 472Mb of RAM and has an existing Drupal installation on it, so I know
I am asking a lot of the box, but my options number one at this point. An alternative place is being negotiated.
I have copied over the previous /var/www/ and when I brought the site up ran git pull and am presently running 3.0.1541 - 1156. I have turned off the Facebook connector so as to simplify matters a little.
If I have extended the cron job interval to 59 minutes as I suspected it would take longer to complete. When I started up, the cron job triggered and I get lots of process of the form:
/usr/bin/php include/onepoll.php 106 https://friend.elsmussols.net
and few of the form:
sh -c '/usr/bin/php' 'include/onepoll.php' '5' 'https://friend.elsmussols.net' &
And load rises, if it hits 50 I have to killall php to calm the box down again. I have not successfully got to the end of the cron process yet I think (you can only stare at htop for so long...).
I have also noticed a few processes that are using https://www.elsmussols.net
which is flat out wrong.
Does onepoll hit every contact every run? If it is interrupted are we back to the bginning every time it runs?
I am seriously looking for some guidance here... I will obviously provide log and output info if required. I need this to go right, I am doubting my ability to do anything techwise at the moment.