On 5/21/26 02:25, Bob Proulx wrote:
Eli Zaretskii wrote:
However, it looks like these problems still happen intermittently.
Just half an hour ago, all the 5 admindb pages of the Emacs mailing
lists timed out, but 5 minutes later they were displayed just fine.
I don't know why but apache2 wedged up on lists.gnu.org and the Monit
utility restarted it.
The problem is one of external influence. The machine is interacting
with the Internet. When things heat up it has an effect.
There is basically only three things that machine does. One is the
exim mail transport agent. One is mailman. One is apache. Those
three services all spawn multiple sub-processes depending upon what
they are doing. If they all become active at the same time it bogs
things down somewhat.
And then the system it is running on is a virtual machine shared among
the 22 virtual machines running on that hardware. If other systems
are active then they all affect each other by using the shared CPU and
shared I/O bandwidth.
Are there other Web servers on the same VM host that come under
sustained MACAC attacks? Could abusive crawlers be tying up, not only
the httpd, but also the VM host and therefore the other services running
in different VMs?
And then the root file system is on the storage area network on other
systems accessed over the network. Sometimes the network on that side
is problematic.
I look at the Uptime Kuma for the site and about the same time there
is a large column of failures across many of the systems. I don't
know why. Something happened.
I know abusive crawlers are the obvious culprit these days, but could
that abuse be covering up another problem, such as growing hardware
flakiness?
-- Jacob
