headers
Steven Levine | 11 May 2013 00:04
Picon
Favicon

Consider setting never_obfuscate_mailto in trac.ini?


Hi Adrian,

It has been brought to my attention that mailto links such as the link at
http://svn.netlabs.org/acpi#MailingList do not display even if the user is
logged in.

It may have been this way for a while, but I didn't notice because I have
TRAC_ADMIN permissions.  Based on the trac docs, I suspect this occurred
during the TRAC update that added a number of email formatting controls.

I can't see any reason that we should not want explicit mailto links in
the wiki content to be displayed.  Unless some of the other admins object,
I recommend we add 

  never_obfuscate_mailto = true

to the global tracini.  This seems like a better solution than giving
authenticated users EMAIL_VIEW permission which would make all email
addresses visible.

Thanks,

Steven

--

-- 
----------------------------------------------------------------------
"Steven Levine" <steve53 <at> earthlink.net>  eCS/Warp/DIY etc.
www.scoug.com www.ecomstation.com
----------------------------------------------------------------------
(Continue reading)

Permalink | Reply | 1 comment |
headers
Adrian Gschwend | 9 May 2013 15:03
Picon

Again server update

Hi group,

Again some updates on the server front, looks like it still traps,
although in a less frequent rate right now.

I analyzed the trap dumps with our fellow kernel coder Knut, seems to be
pretty hard to figure out what exactly goes wrong but it's almost
certainly memory related. As they replaced the blade it is most probably
either CPU or the memory itself.

I will talk to the ISP to see if they can replace both, which would
probably be the best bet.

Not sure how long this takes tough, will keep you posted.

regards

Adrian

--

-- 
Adrian Gschwend
 <at>  netlabs.org

ktk [a t] netlabs.org
-------
Open Source Project
http://www.netlabs.org
Permalink | Reply | 2 comments |
headers
Adrian Gschwend | 6 May 2013 17:16
Picon

Server issues

Hi all,

As you noticed the server was down again, the new hardware trapped on
Sunday and it was not easy to get it back up.

Now it just trapped again, looks like there are issues with the brand
new CPU ("Machine check exception" is usually not a good sign).

I will work with the ISP which hosts the machine to figure out what is
going wrong.

sorry for the troubles!

regards

Adrian

--

-- 
Adrian Gschwend
 <at>  netlabs.org

ktk [a t] netlabs.org
-------
Open Source Project
http://www.netlabs.org
Permalink | Reply | 3 comments |
headers
Lars Erdmann | 4 May 2013 21:19
Picon
Favicon

Netlabs not accessible

Hallo Adrian,

SVN does not work, neither for USB nor for ACPI. Trac won't work either. 
The web page won't show up.
Permalink | Reply | 1 comment |
headers
Adrian Gschwend | 12 Apr 2013 13:54
Picon

TRAC notification

Hi group,

ok that's a bit embarassing but I figured out why TRAC notification
didn't work on some projects: I didn't enable it by default, it's only
there when it's configured for the project.

So I wonder what I should do, do the project owners tell me and I enable
it or should we enable it by default for all of them?

Also it will still require people to register their email address
because TRAC cannot figure that out from LDAP.

Thoughts?

Thanks to Andy for the hint :-)

cu

Adrian

--

-- 
Adrian Gschwend
 <at>  netlabs.org

ktk [a t] netlabs.org
-------
Open Source Project
http://www.netlabs.org
(Continue reading)

Permalink | Reply | 6 comments |
headers
Adrian Gschwend | 28 Mar 2013 18:47
Picon

Services migrated

Hi group,

It looks like I could successfully migrate the remaining services to the
new server. The services moved today were SVN, TRAC, Jabber/XMPP & SIP.

DNS propagation is pretty much done, you should be able to access all
services. If this is not the case, please let me know.

In the next days I will update SVN & TRAC as well so expect some more
maintenance but there should not be long downtime anymore.

I've tested mail notification on TRAC and it worked for me, if you
really have issues with that let me know.

There are a few things left like the strange problem with the updated
SVN & TRAC SSL Certificate which I will debug in the next days. Also the
Piwik server refuses to start but that should not be a show stopper
right now.

Yesterday we had 4 hours downtime on the new server as the KVM switch
was broken in the (new) blade server, they replaced the box and now
everything seems to work fine.

If I missed anything let me know.

cu

Adrian

--

--
(Continue reading)

Permalink | Reply | 14 comments |
headers
Adrian Gschwend | 18 Mar 2013 20:54
Picon

Downtime Tuesday morning 6:00 CET

Hi guys,

Tomorrow Tuesday morning there is a power outrage at one of the
netlabs.org server locations. I wanted to migrate services to the new
server already but couldn't manage to fix all open issues in time.

So we will be hit by the downtime, services will go down around 6:00 CET

Mapping to different time zones:
http://everytimezone.com/#2013-3-18,1020,6bj

Will put the machine back up as soon as we have power again (and I'm
awake ;)

regards

Adrian

--

-- 
Adrian Gschwend
 <at>  netlabs.org

ktk [a t] netlabs.org
-------
Open Source Project
http://www.netlabs.org
Permalink | Reply | 0 comments |
headers
Lars Erdmann | 18 Mar 2013 20:22
Picon
Favicon

Netlabs TRAC: odd behaviour

Hallo Adrian,

whenever anyone adds anything to a ticket (the bug reporter or myself), 
this error pops up:

"an error occurred while sending notifications: [Errno 8] hostname nor 
servname provided, or not known"

This has only happened for the last week or so. The ticket will be 
eventually updated but it takes quite some time. Does the user need to 
do anything to fix this behaviour or does TRAC need a change ?

Lars
Permalink | Reply | 1 comment |
headers
David Azarewicz | 16 Mar 2013 17:34
Favicon

Certificate error

Hello,

I'm getting this error when trying to do an SVN commit.

Error validating server certificate for 'https://svn.ecomstation.nl:443':
 - The certificate is not issued by a trusted authority. Use the
   fingerprint to validate the certificate manually!
 - The certificate hostname does not match.
Certificate information:
 - Hostname: trac.netlabs.org
 - Valid: from Tue, 12 Mar 2013 11:21:51 GMT until Thu, 12 Mar 2015 11:21:51 GMT

 - Issuer: http://www.CAcert.org, CAcert Inc.
 - Fingerprint: 26:43:e1:ab:21:5f:39:03:e7:52:e9:99:a0:81:6f:60:86:37:04:53

Is this OK?

Thanks,
David
-----
David Azarewicz
http://www.88watts.net
Permalink | Reply | 2 comments |
headers
Adrian Gschwend | 10 Mar 2013 23:44
Picon

Server migration

Hi everyone,

I'm currently migrating netlabs.org services to the new hardware. This
is done in steps, first I virtualize the existing installations and then
I will split them to new (virtual) machines.

To do that I also have to change the IP addresses as I got a new network
assigned from the provider.

So in the next days there is a fair chance you will experience
interruptions of services due to taking down the old system and
activating the new one. Also as usual it takes a while to synchronize
DNS servers, especially the secondary ones.

Will keep you posted when I'm done with the basics.

cu

Adrian

--

-- 
Adrian Gschwend
 <at>  netlabs.org

ktk [a t] netlabs.org
-------
Open Source Project
http://www.netlabs.org
(Continue reading)

Permalink | Reply | 1 comment |
headers
Adrian Gschwend | 2 Mar 2013 12:49
Picon

Problems with SVN server

Hi group,

As some of you pointed out last week there are some issues with the SVN
server. Last week the box locked itself three times within less than 24
hours.

Unfortunately I couldn't analyze what is going on, the box was still
alive (in terms of the kernel was ok) but I couldn't get a console
anymore. That's usually related to filesystem issues.

After a reboot everything looked good, the filesystem integrity was ok
too. Analyzing stats I noticed that I had a lot of open file handlers
when the issues started, in the range of > 10'000 while normal load is
more around 1'000 in average.

This happened again twice on the same day, unfortunately not during work
hours so I didn't notice it first.

After that it behaved ok for some day, everything looked just as
expected. Now last night it happened again and fortunately I could at
least check which processes are going wrong. It indeed looks like SVN is
the issue, I get a *lot* of open file handlers related to Apache processes.

After stopping the jail with SVN in it the server is back to normal.
Will wait a bit to let the FS calm down and restart it again after that.
What exactly goes wrong I don't know yet.

Anyway as I wrote last week there is a powerful new server waiting to
replace the current box. I will start working on that next week so the
issues are hopefully solved soon.
(Continue reading)

Permalink | Reply | 0 comments |

Gmane