Toolserver-l August 2009

toolserver-l@lists.wikimedia.org

48 participants
32 discussions

[important] conversion of wolfsbane (web server) to Solaris
by River Tarnell 18 Aug '09

18 Aug '09

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hi, following the implementation of the Solaris login server (willow), we are planning to move the web server, wolfsbane, to this platform. compared to the previous change on the web server (from Apache to ZWS), we don't expect this to require as much effort from users. for example, there is no need to change your htaccess or rewrite rules, because we will be using the same web server software. for most tools (e.g. simple CGI or PHP scripts), there will be no need to change anything. in order of likelihood, the most common problems will be: * missing modules (for PHP/Perl/Python) * missing libraries (for C/C++) * different versions (e.g. PHP 5.3 vs 5.2, Python 2.6 vs 2.5) * OS differences if you find missing modules or libraries, please file a request on JIRA to have these installed. for version differences, normally no changes will be required; if any are required, they will be documented in the release notes for the language in question, and should be easy to do. in extreme cases (where converting to the new version would be very difficult), we may be able to install old versions of the affected software. there are unlikely to be many people affected by OS differences; this would only happen if your tool calls external programs, which is uncommon. if you encounter a problem of this sort and can't solve it yourself, you can ask for help on IRC or the mailing list. there are some notes about Linux->Solaris migration on the TS wiki at <https://wiki.toolserver.org/view/Using_Solaris>. there will be a one month test / migration period, during which you can test the web server on willow. to do this, replace "toolserver.org" in your tool's URL with "willow.toolserver.org", for example: http://willow.toolserver.org/~jsmith/mytool.php this will last until Wednesday, 19 August, at which point wolfsbane will be converted. - river. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (HP-UX) iEYEARECAAYFAkpolUwACgkQIXd7fCuc5vL2EgCgmG45stK/gAuTQS48mSAumkPL svkAnRV66vax/6j+Mrq5E/1/YsF4ESoq =lwH6 -----END PGP SIGNATURE-----

1 2

RfC: globalsitenotice
by DaB. 18 Aug '09

18 Aug '09

Hello all, at the moment it is difficult to inform our toolusers (people that use our tools) about planed maintainces, downtimes and other important things. So I had the idea today that a kind of global sitenotice (like the one in the wikimedia-projects) would be nice and a easy solution for that. The easiest way to implement it is IMHO a simple textfile, that toolauthors may/can/should/whatever insert in their tools. I created such a file, it can be found at /var/www/sitenotice on wolfsbane and via http://toolserver.org/sitenotice. For normal the file would be empty or contains only a empty div-container. If the roots know about a planed downtime or an important change, they would insert a short text in the file and our toolusers would be informed. The setup is differend to status.toolserver.org, because status lists only the current status, not a futuraly status. Please comment at the toolserver-mailinglist and tell me/us if you would implement the sitenotice in your tools. Sincerly, DaB. -- wp-blog.de

15 21

Re: [Toolserver-l] commonswiki status
by DaB. 18 Aug '09

18 Aug '09

Hello, Am Montag 17 August 2009 21:44:28 schrieb Carl Fürstenberg: > Perhaps there would be an idea to initiate an other dev to be able to > do such work. it is not this easy as you think. For taking a dump, a db-slave of the wm- cluster has to taken away from the cluster. That means that the cluster miss one server and the other servers has to do more work. The new commons(=S4)- cluster have fewer server then the other cluster so taken one away is even more critical. So such a old could be only taken in low-traffic-times. Sincerly, DaB.

3 8

Unplaned rebooting of linux-servers
by DaB. 15 Aug '09

15 Aug '09

Hello all, to make sure that CVE-2009-2692 is fixed after the kernel-update, I rebooted all linux-servers (that's nightshade, wolfsbane and cassini) tonight. The downtime was fewer then 5 minutes. Sincerly, DaB. -- wp-blog.de

1 0

queued mail from willow
by River Tarnell 09 Aug '09

09 Aug '09

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hi, people running cron jobs or otherwise receiving system mail on willow will not have received that mail in the last few days, as it was being queued instead of sent. this has now been fixed, so you may see some old mail arriving. - river. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (HP-UX) iEYEARECAAYFAkp+gW8ACgkQIXd7fCuc5vKdRwCgqPpJ2VPsihkQfqoDuvuEYnT2 NoEAn23buSxsRd1cqeXoswqGsmHrhJV9 =etIX -----END PGP SIGNATURE-----

1 0

s1 status
by River Tarnell 09 Aug '09

09 Aug '09

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hi, the s1 server (rosemary) has a faulty motherboard which is causing system crashes. we will replace the part next week; until then, access to s1 will be intermittently offline. - river. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (HP-UX) iEYEARECAAYFAkp+CUkACgkQIXd7fCuc5vJPoACbB3A2wdatbZvrdwSUQ1bcm3wH eT4AnA0mKf1cghr/a6FPRoIvGmOIqYmb =oi+H -----END PGP SIGNATURE-----

4 6

s1 maintenance Saturday morning
by River Tarnell 07 Aug '09

07 Aug '09

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 hi, since the s1 import, replication has been unacceptably slow. unless this improves in the next few hours, i will do some additional maintenance tomorrow morning which should help. this will last several hours. - river. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (HP-UX) iEYEARECAAYFAkp8ZlgACgkQIXd7fCuc5vIkpgCfQS62J0WqeO3EwBEbMMO96veW IxAAoKptxbhROfLJb7G34yRAjyHDB3X4 =mrtj -----END PGP SIGNATURE-----

1 0

JIRA and Fisheye down?
by James R. 06 Aug '09

06 Aug '09

I was just trying to access https://jira.toolserver.org/ and https://fisheye.toolserver.org/ and I'm getting a "problem loading page: the server at jira.toolserver.org is taking too long to respond" error. Is there any planned downtime which I was unaware of? Thanks. -- Regards, JamesR English Wikipedia Administrator http://en.wikipedia.org/wiki/User:JamesR

2 1

Unplaned reboot of nightshade
by DaB. 06 Aug '09

06 Aug '09

Hello all, I had to reboot nightshade unplaned, because I broke the /dev and no login where possible anymore (I guess it broke many other things too). Users told me, that the reboot-message was not received too, sorry again. Sinclery, DaB. -- wp-blog.de

3 2

Re: [Toolserver-l] where's enwiki_p on s1?
by Lilewyn 06 Aug '09

06 Aug '09

That was my not looking first. <s>MZMcBride</s>Another, to be unnamed, toolserver user just informed me that there was such a text box on the page. I mentioned to the channel this conversation was in that I made the note while waiting for status.toolserver.org to load (it still hasn't, so presumably the issue is local). Sarcasm would be when it was suggested to me that I see if there's a "status.status.toolserver.org" while I wait. ~ Kylu Date: Wed, 5 Aug 2009 23:47:41 +0200 From: "DaB." <WP(a)daniel.baur4.info> Subject: Re: [Toolserver-l] where's enwiki_p on s1? To: toolserver-l(a)lists.wikimedia.org Message-ID: <200908052347.44221.WP(a)daniel.baur4.info> Content-Type: text/plain; charset="iso-8859-1" Hello, Am Mittwoch 05 August 2009 23:43:03 schrieb Lilewyn: > I'm curious, would it be worthwhile to add a tsadmin-editable textbox to > status.toolserver.org, that way the toolserver admins could just point at > it in the future, and we users would eventually learn to check there for > service ETAs instead of asking directly? is that sarcasm? Sincerly, DaB. -- wp-blog.de __________________________________________________________________ Looking for the perfect gift? Give the gift of Flickr! http://www.flickr.com/gift/

2 1

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Toolserver-l August 2009