NOTE FROM SUPERSLEEPER:
[for most recent updates, scroll to the
bottom of the page]
SERVER
UPGRADE INFO:
UPDATE:
March 16, 2020 2:45 PM:
The server updates have now been completed.
We are now running with the upgrades. That said, there is still
an off-again, off-again DoS attack that is ongoing. The reason
we upgraded the server was to better address and handle situations
like this DoS attack with more RAM & faster processor, so that
although there may be some degree of slowness from time to time, the
website & forum will hopefully not have to shut down as a
result. Thanks to all our forum members for your support and
your patience.
If you would like to help with the monthly costs
associated with maintaining the server, please consider a donation to
help us out, at this link:
Thanks folks!
OLDER UPDATES:
2:30 PM (Central Time, US), March 14, 2020.
Howdy folks,
The datacenter that houses our server was having some issues all this
morning starting at about 9 am or thereabouts, U.S. Central Time.
They are aware of the problem and are currently addressing it. The
server loads have come considerably down from their highs earlier this
morning, but these issues are coming back still as I write this.
In the meantime, I'm keeping the forum area closed (for now). All
other areas, including the Wiki, the main website, the Mantis site (for
the OSCAR development team) and the OSCAR download site will remain open,
but may be slow in response time. I'm not going to re-open the forum
just to have to close it again, so I want to first make sure that the
loads will be reduced for a while, so I don't open things up and then have
loads spike again 20 minutes later. So allow me some time on this -
I want to make sure that our forum members have the system fully working
well before re-opening the forum.
This temporary
measure should no longer be needed once the datacenter assures me that all
is well again.
After this particular issue is completely resolved, I'm going to put in an
order for an upgrade to our server, which will likely double our RAM &
equip us with a faster processor so that we can absorb issues like this
without having to shut down the forum. We don't really have the
funds to do that, but we'll figure all that out later.
In the past, Apnea Board has remained online with something like a 99.8%
uptime, so this particular issue (and the shorter one we had last week)
are highly infrequent.
I apologize for the inconvenience, folks. I'm trying my best to make
sure we have a robust system for the future and I'm spending most of my
day today keeping on top of this, and doing little else.
I will provide updates on this page as
I learn anything new. (below this note, each with a time-stamp).
{might have
to re-load this page to see the updates}
Thanks,
SuperSleeper
Apnea Board Administrator
www.ApneaBoard.com
_____________________________
UPDATE 2:43 PM:
Server loads seem back down to near
normal levels, but forum will remain closed until we're absolutely sure
they'll stay that way. Nothing new from the datacenter at this
point, but it's starting to look like they're having some success.
Will provide updates as they come.
SS
____________________________
UPDATE 3:01 PM:
Server loads continue to remain
normal. If this low level continues, I anticipate opening the forum
back up in about 15 minutes, around 3:15 PM Central Time (US). If
loads go back up, we'll hold off a bit more. Updates will be posted
here.
____________________________
UPDATE 3:10 PM:
Server just got a huge spike. So
I'm backing off my 3:15 estimate for a bit. Sorry.
_____________________________
UPDATE 3:37 PM:
Datacenter Shift Supervisor states the
following "We do apologize for the issues you've been facing recently,
our senior admins are aware and are working to correct this.
During this time you may noticed fluctuating load averages but we are
doing our best to minimize any disruption to your services."
Although we're still getting smaller
spikes, I'm keeping the forum closed until we've gone about 30 minutes
with no major load spikes. Will continue to update you, next one
shortly after 4:10 PM hopefully.
_____________________________
UPDATE: 4:10 PM:
The datacenter people did some sort of
updates/patches or replaced some network equipment (I have no clue) and
then re-booted the server, which took about 10 minutes. No one could
access the website during that re-boot time. It appears to be up and
running now. Still will be waiting now until 4:40 pm Central Time
(US) to make sure all is well. Will keep monitoring server load and
will re-assess opening the forum back up at that time, if the load remains
low for that 30-minute period.
____________________________
UPDATE: 5:08 PM:
The load has been relatively low
now for well over 30 minutes. I'm going to open the forum back up
and we'll see how things go. We are still having smaller load spikes
that seem to last for only 30 seconds or so, so you will undoubtedly see
occasional slowness or even time-outs. When that happens, just
come back 5 minutes later and it should be back down to normal
levels. As always, if we have to take the forum back down for a bit,
I'll post here (assuming the server is not completely down).
Please read my thread in the Main Forum
here:
That thread contains the following
advice:
SUGGESTION: If you're typing a rather
large forum post, I'd suggest copying
your text PRIOR to hitting the "Post Reply" or "Post Thread"
buttons, just in case we have to shut down the forum again right as
you are submitting your post. That way, you can save the text of
your post in a text document (so that your work is not lost) and then
paste it into a post later when the forum comes back online.
____________________________
____________________________
UPDATE:
March 15, 2020 2:25 AM:
The
server has been up and running without any major load spikes
since 5 PM Saturday evening. So far so good. The forum and
everything else remains open and running okay.
____________________________
UPDATE:
March 15, 2020 12:45 PM:
Server
load spikes have returned to some degree, although not quite as bad as
yesterday, and the spikes seemed to be mitigated so that they only
last for about 3 minutes, then things seem to return to more normal
levels. This new action is probably due to some of the
mitigating tweaks the datacenter did yesterday, which may help us to
remain online with no major issues. That said, we may have to
take the forum offline for brief periods if it gets worse. Am in
contact with the datacenter again. Will update when possible.
___________________________
UPDATE: March
15, 2020 4:03 PM:
The
intermittent issues still persist. The behavior seems to be
everything normal for several minutes, then high load spikes for 3-5
minutes now, then back to normal. The datacenter shift
supervisor is escalating these issues to a higher level admin.
Sounds like they might be doing something with network equipment or
something. Our server may be disconnected from the Internet at
some point during the change, but it probably won't be for more than a
few minutes or so.
___________________________
___________________________
UPDATE:
March 16, 2020 1:40 PM:
We
are in the process of upgrading the server. The forum is
currently shut down, and we will be taking the server off line for
upgrades, and then a re-boot, and hopefully on the other end, we'll
have a faster, more robust system. Thanks for your
patience!
____________________________
UPDATE:
March 16, 2020 2:45 PM:
The
server updates have now been completed. We are now running with
the upgrades. That said, there is still an off-again, off-again
DoS attack that is ongoing. The reason we upgraded the server
was to better address and handle situations like this DoS attack with
more RAM & faster processor, so that although there may be some
degree of slowness from time to time, the website & forum will
hopefully not have to shut down as a result. Thanks to all our
forum members for your support and your patience.
If
you would like to help with the monthly costs associated with
maintaining the server, please consider a donation to help us out, at
this link:
Thanks
folks!