Not signed in

Want to take part in these discussions? Sign in if you have an account, or apply for one below

Site Tag Cloud

Vanilla 1.1.10 is a product of Lussumo. More Information: Documentation, Community Support.

• CommentRowNumber1.
• CommentAuthorUrs
• CommentTimeSep 11th 2010
• (edited Sep 11th 2010)

looks like the lab is down. I get “Internal error” when calling any page.

I can’t do anything about it, certainly not from where I am right now. Hopefully Toby can find a minute to restart the server or something.

• CommentRowNumber2.
• CommentAuthorTim_Porter
• CommentTimeSep 11th 2010

In case anyone wonders, it seems now to be working.

• CommentRowNumber3.
• CommentAuthorUrs
• CommentTimeSep 11th 2010
• (edited Sep 11th 2010)

Yes, works for me, too. Did nobody else notice anything? Strange.

It happened to me while I was editing infinity-Chern-Weil theory introduction. That entry, like similarly long entries, anyway and usually takes a loong time to re-display after one edits it. The time spans were getting ever longer last night. Then finally it got to the point that after five minutes or so of waiting, the page would disappear entirely and just say “Internal error”. Then I tried any other page, and all just gave me “internal errors”.

• CommentRowNumber4.
• CommentAuthorEric
• CommentTimeSep 11th 2010

I also noticed it was down, but after seeing your post here, didn’t speak up about it.

• CommentRowNumber5.
• CommentAuthorUrs
• CommentTimeSep 11th 2010
• (edited Sep 11th 2010)

I also noticed it was down

Thanks for saying this. I was seriously starting to believe I was suffering from hallucinations. (After all it was around 4am here, after a long day…)

• CommentRowNumber6.
• CommentAuthorUrs
• CommentTimeSep 11th 2010
• (edited Sep 11th 2010)

Maybe i am wrong, but my impression is that loading lengthy pages with lots of links throws the server into a hugey inefficient search for whatever it thinks it must search for.

Try to request two or three of the lengthier pages at once. If it works on my system, it will take the order of a minute for these to appear, tendency rising. Now at the same time, while these are loading, open a mini-page, such as the HomePage. Check if that also appears very slowly now. It seems to me it does. Which would mean that the server is really busy doing something it thinks is important, which however it shouldn’t be doing.

• CommentRowNumber7.
• CommentAuthorUrs
• CommentTimeSep 12th 2010

is this just on my end, or is the nlab immensely slow currently?

• CommentRowNumber8.
• CommentAuthorUrs
• CommentTimeSep 12th 2010

… or is it actually down?

(When a reaction takes minutes, do we say “it was down and came back” or just “it’s slow”? ;-)

• CommentRowNumber9.
• CommentAuthorTobyBartels
• CommentTimeSep 12th 2010

I just did a hard restart, so it’s back.

• CommentRowNumber10.
• CommentAuthorRodMcGuire
• CommentTimeSep 12th 2010

all my attempts to load a page hang. (one has been hanging for 25 minutes as has a save page attempt)

• CommentRowNumber11.
• CommentAuthorTobyBartels
• CommentTimeSep 12th 2010
• (edited Sep 12th 2010)

I’m editing now without trouble.

Rod: Try saving your content in an external editor (Notepad, emacs, etc) then reload the page and submit again.

• CommentRowNumber12.
• CommentAuthorUrs
• CommentTimeFeb 9th 2011

I am trying to restart the Lab… but currently I seem to be failing. Not sure what’s going on.

• CommentRowNumber13.
• CommentAuthorUrs
• CommentTimeFeb 9th 2011

it has decided to come back to us after all

• CommentRowNumber14.
• CommentAuthorzskoda
• CommentTimeFeb 11th 2011
• (edited Feb 11th 2011)

Some $n$Lab accessibility time out at the moment (20:12 CET) Edit: get something about a minute or two for a 2 line entry. Edit (20:40): the $n$Lab is back. Edit (20:42): it lasted for short…

• CommentRowNumber15.
• CommentAuthorUrs
• CommentTimeApr 16th 2012
• (edited Apr 16th 2012)

The lab has been very slow the whole day. But now it has not been reacting for more than 60 minutes.

• CommentRowNumber16.
• CommentAuthorAndrew Stacey
• CommentTimeApr 16th 2012

Now that’s one I haven’t seen before. The nLab processes were using up the whole CPU allocation and seemingly not releasing it. I tried a soft restart and it didn’t take any notice of that, so I killed the nlab processes and am watching it to see if it blows up again.

• CommentRowNumber17.
• CommentAuthorMike Shulman
• CommentTimeApr 16th 2012

I can’t get to it at all right now, but that could be partly my connection (hotel room).

• CommentRowNumber18.
• CommentAuthorUrs
• CommentTimeApr 16th 2012
• (edited Apr 16th 2012)

Thanks, Andrew, for looking into the problem!

It’s still not back, though, is it? Or it “blew up” right away again, or something. In any case, it doesn’t respond.

[edit: I see I overlapped with Mike.]

• CommentRowNumber19.
• CommentAuthorUrs
• CommentTimeApr 16th 2012

now it seems to be back

• CommentRowNumber20.
• CommentAuthorTobyBartels
• CommentTimeApr 25th 2012

Extremely slow response times now.

• CommentRowNumber21.
• CommentAuthorTim_Porter
• CommentTimeApr 25th 2012

I am getting no response either.

• CommentRowNumber22.
• CommentAuthorTobyBartels
• CommentTimeApr 25th 2012

Now it’s fine!

• CommentRowNumber23.
• CommentAuthorUrs
• CommentTimeMay 11th 2012

Today and the last days, on my end the $n$Lab has been quasi-down frequently. All along when I am working on it, I have to stop and do something else for 10 minutes here and 10 minutes there, to wait for the Lab to come back.

I might be the only person experiencing this. But since it’s not getting better, I am wondering if maybe some restart or something might help. (Maybe some dead processes have been accumulating or something? Of course I don’t know.)

• CommentRowNumber24.
• CommentAuthorMike Shulman
• CommentTimeMay 11th 2012

I’ve noticed it once or twice. (I haven’t been using the lab very heavily though.)

• CommentRowNumber25.
• CommentAuthorTobyBartels
• CommentTimeMay 11th 2012

I haven’t noticed that lately, but I have noticed such things happening from time to time. I only report here if I can’t get anything for maybe a quarter of an hour.

• CommentRowNumber26.
• CommentAuthorTobyBartels
• CommentTimeMay 11th 2012

Although I can’t get it right now, it turns out.

• CommentRowNumber27.
• CommentAuthorFinnLawler
• CommentTimeMay 11th 2012

The lab hasn’t been responding for me at all this evening (or yesterday, come to think of it).

• CommentRowNumber28.
• CommentAuthorUrs
• CommentTimeMay 11th 2012

I have these phases where I use the nLab very intensively as my tool to write seminar notes. In these periods, I notice it heavily that it is down a lot. I am not announcing this here so much, as in the past is had led to some bad feelings. Sometimes I wonder if many edits per hour are actually the cause of the problem, not just a way of noticing it.

• CommentRowNumber29.
• CommentAuthorUrs
• CommentTimeMay 12th 2012
• (edited May 12th 2012)

Here is maybe an interesting data point:

in these quasi-down phases, such as now, it frequently happens to me that after the lab did not respond for a while, it suddenly comes back on a page for which I requested to edit it with the message

“entry is locked, Urs Schreiber has been editing this page for 29 minutes”

where in fact I have been staring at a blank screen for 29 minutes (after hitting edit twice, apprently).

So apparently the part of the system that recognizes if an edit is requested is still alive, while only the part that should serve up the content is dead.

(Trouble is of course that this behaviour makes it all the more frustrating, when the request to break the “lock” results in another 29 minutes of waiting…)

• CommentRowNumber30.
• CommentAuthorTobyBartels
• CommentTimeMay 13th 2012

I’ve had that on occasions where I’m confident that I did not hit edit twice (or reload the page). However, possibly my browser repeated the request (which is GET, not POST).

• CommentRowNumber31.
• CommentAuthorDavidRoberts
• CommentTimeMay 22nd 2012

The lab seems to be down for me.

• CommentRowNumber32.
• CommentAuthorUrs
• CommentTimeMay 22nd 2012
• (edited May 22nd 2012)

Try again now. In the last weeks the $n$Lab becomes unresponsive, I would estimate, for around 15-30 minutes every 1.5 hours or so, on average.

• CommentRowNumber33.
• CommentAuthorDavidRoberts
• CommentTimeMay 22nd 2012

Yeah, it came back in roughly that time span and I forgot to update :S

• CommentRowNumber34.
• CommentAuthorAndrew Stacey
• CommentTimeMay 22nd 2012

I’ve just killed some “rogue processes” on the server. These were instances of instiki that had somehow gotten “detached” from the web server and were sitting eating up the CPU. They’ve been hanging around for a few weeks, it seems. I’m hopeful that killing these will stop the bottlenecks that have been happening.

(To be clear, they haven’t been doing anything malicious other than using up resources.)

• CommentRowNumber35.
• CommentAuthorUrs
• CommentTimeMay 22nd 2012

Thanks, Andrew!!

• CommentRowNumber36.
• CommentAuthorAndrew Stacey
• CommentTimeMay 23rd 2012

Do let me know if this fixes the slow-downs.

• CommentRowNumber37.
• CommentAuthorUrs
• CommentTimeFeb 12th 2013
• (edited Feb 12th 2013)

Something is bothering the $n$Lab theses days. It has been slow (though responsive throughout) through the last two weeks or so.

But today, it is getting to the point of not coiming back anymore. I have found it justified to do a restart three times today. The good message is that the restart did have a positive effect – for a while.

Just FYI. In case it is of any help for anything.

• CommentRowNumber38.
• CommentAuthorTodd_Trimble
• CommentTimeFeb 12th 2013

Would it bother the nLab for a user to have multiple windows open on different pages? (I am really ignorant about such things.) None of the windows I have open is in edit mode at the moment.

• CommentRowNumber39.
• CommentAuthorAndrew Stacey
• CommentTimeFeb 12th 2013

Todd, it shouldn’t do. The only time it might cause a bottleneck is if you tried to reload all of the pages simultaneously.

Urs, I’m not aware of any current issues but I’ll look into it.

• CommentRowNumber40.
• CommentAuthorUrs
• CommentTimeApr 8th 2013

The lab is down, and I can’t bring it back.

When I log into the server and ask it to restart, it claims that it’s “Unable to connect to the system bus…”

• CommentRowNumber41.
• CommentAuthorAndrew Stacey
• CommentTimeApr 8th 2013

It’s back up now. An error like that suggests a problem with the host. My guess is that they detected it and fixed it.