-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
hi,
on Monday 14th, around 2AM UTC, we will do further maintenance on the
sql server (zedler) to move the databases back to the original server.
based on the amount of data to copy, i expect the downtime to be around
2.5 hours.
- river.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (HP-UX)
iEYEARECAAYFAkqpMtIACgkQIXd7fCuc5vJmlwCggzc0M3/MuEa3pKT668AyZid3
8hkAoL+5CdBEQjs7Ua6+R3nKsI5BjI02
=PQWi
-----END PGP SIGNATURE-----
Hello there. This is a quick status update on the project to get
OpenStreetMap maps into Wikipedia and on the Wikimedia/OpenStreetMap
toolserver setup. For those that don't know what it's all about we're:
* Setting up a testing platform (ptolemy & ortelius) to serve OSM
maps in Wikipedia which can be rolled into production
* Setting up a toolserver (cassini) which interested parties can use
to write tools that use OSM data. And combine it with Wikimedia data
if they want.
Here's an (outdated!) wiki page with some more info:
http://meta.wikimedia.org/wiki/OpenStreetMap
And here's a recent talk I gave at Wikimania discussing the current status:
http://wikimania2009.wikimedia.org/wiki/Proceedings:143
Until now we only had 1/3 machines active & accessible due to various
combinations of waiting for hardware, people being busy and it being
unclear who actually got access to those machines that I won't go
into.
That one machine was the nascent WMOSM Toolserver Cassini. I set up a
prototype multilingular rendering (since ptolemy and ortelius weren't
available at the time) which you can see here:
http://cassini.toolserver.org/tile-browse/
What we need to to currently is:
* Admins to set up ptolemy / ortelius so that they replicate the DB /
mirror tiles
* Get interested users/developers to *use* cassini for their tools so
we can get neat stuff like the multilingular-country-list
(http://cassini.toolserver.org/~mazder/multilingual-country-list/).
Sign up here: https://wiki.toolserver.org/view/Account_request
* Cassini also needs some admin love
* Work on this buglist (and more bugs) to get OSM into Wikipedia:
http://tinyurl.com/nv2sap
I did most of this on Cassini (see setup notes:
http://tinyurl.com/mbblj3) but I have limited time on it (especially
until Christmas) and system administration isn't my strong suit. So
unless we get other people to help this project is going to go
*slooowly* and you won't have OSM on Wikipedia until 2010.
Before WM2009 we had the unfortunate problem of interested parties
needing to do the above not getting access due to the issues above.
However the machines are up *now* and during WM2009 uncertainties
about who could grant access were finally solved (brion will be
dictator).
So, any potential admins for ptolemy and ortelius will have to:
1. Be qualified & motivated
Preferably someone who's worked with the OSM toolchain or is willing
to learn. If you're maintaining your own ad-hoc rendering somewhere
and are running out of server space you'll probably be motivated to
get this working sooner.
2. Reveal their name & address to the Wikimedia foundation
That's a requirement Wikimedia requires of all server administrators.
3. Agree to Wikimedia's privacy policy
URL: http://wikimediafoundation.org/wiki/Privacy_policy
ptolemy and ortelius are otherwise distinct servers on the Wikimedia
network so this amounts to basically not being silly and running a
Quake III server on them, or nmap-ing the Internet.
Cassini is more sensitive and harder to gain access to since root
admins on Cassini have access to private data about Wikimedia users
(e.g. raw database access, login cookies and so on).
But in both cases we should be able to give elevated non-Unix-root
privileges. All of this pending approval by Wikimedia of course.
I was trying to download data from http://toolserver.org/~kolossos/templatetiger/ using a download manager, but suddenly the following problem occurs
Can't connect to MySQL server on '0' (146)
I'm extremely sorry if it is done by me. plz help me to access it again.
By the way, I also requested for a query service at
https://jira.toolserver.org/browse/DBQ-73
and nobody has yet responded. What should I do?
Thanks
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
hi,
the user database server (zedler) has run out of space on the MySQL
filesystem; however, there is still 250GB free in the InnoDB table
space. to reclaim this free space, we need to dump the reload the
database. the database will be inaccessible while this is in progress.
based on the time taken to run nightly backups, the dump should take
about 5 hours. however, i will run it in parallel to improve
performance. it will take a bit longer to reload it afterwards.
after the reload, i will not restore the commonswiki replica that was
previously on the sql server. the correct way to join user databases to
commons is to create them on the sql-s4 server.
- river.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (HP-UX)
iEYEARECAAYFAkqmQNIACgkQIXd7fCuc5vKa/QCfUvvQrs8XaNiztiUTxrS8xA26
0lkAoIpHExWtaamKeKAkXv3YplMnx+zL
=Ivmb
-----END PGP SIGNATURE-----
Hello,
When I upload a file the owner of the file are my user. Now I want to change
the user to the user of a multi-maintainer-project. When I run "chown
commonshelper2 <FILE>" I get the error: "chown: changing ownership of
`<FILE>': Operation not permitted". What do I wrong (I test it with both
user and chmod 777)
Viele Grüße
Jan
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hello,
I have a multi-maintainer account for which I'd like to use
rewrite.script. [1] says to put this file at $HOME/rewrite.script
however it seems to be having no effect. For this account,
$HOME/rewrite.script is /home/project/s/t/e/stewardbots/rewrite.script
- -- is it possible that ZWS is looking at
/home/stewardbots/rewrite.script or something?
Thanks,
- -Mike
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
iEYEARECAAYFAkqoQxgACgkQst0AR/DaKHsD/QCfY2Xg96k9F6U+u5hmWkEQCKYx
ShcAniUh738rFqrMbFnwH1Bh/ja9sEj2
=LeCW
-----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi all
Since the user database is still importing, we decided to at least get the most
important data from zedler back online: the toolserver (meta)database, with
toolserver.wiki etc. To access it, use the hostname sql-toolserver. This way, it
is no longer bound to the locatiuon of the user databases, and we can point that
alias to a workign copy at will. the command line script "sql" was already
changed to do use the new ost name.
So, to fix your tools, change the database host for the toolserver db to
"sql-toolserver".
- -- daniel
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
iEYEARECAAYFAkqnihEACgkQG0czJgig8U9MtwCdHPbJEP2mCprlxVbbOgOoOb+H
NzoAn3AoHkmTau3y287vMYG6s/aqrgGv
=lIin
-----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
hi,
i've unmounted /mnt/user-store in order to convert hemlock from ZFS to
VxFS. this involves copying off the data, reformatting the disk, then
copying the data back; i estimate it will take around 12 hours in total,
during which user-store will be unavailable.
- river.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (HP-UX)
iEYEARECAAYFAkqk3lkACgkQIXd7fCuc5vJz+ACgodImzIzJrqIavmf8E/SoS1QW
BfEAn0yeyexjWw/UTGHVJvDLLZ3u8LkG
=eDpb
-----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hello,
I wanted to set up rsync to backup my home directory to my own machine,
and in doing so discovered that I couldn't read some files in the
stewardbots multi-maintainer account because they had group 5000 (which
I'm not in) instead of group stewardbots (which I am in).
I initially thought this might be done by phoenix, however the .pid
files are not affected (only the .out files), so I don't think that's
the case. Thanks to Simetrical for assistance and helping me think this
through.
Instead, I believe this is caused by a script run from cron to trim the
log files. The command used there was "tail -n $lines $file > $file.temp
&& rm $file && mv $file.temp $file" and I've added to the end a "&&
chgrp stewardbots $file" which I hope will correct the issue.
Does anyone know why that would have changed the file's group (really it
is not changed since the original is deleted, and replaced)? If so, how
can I avoid that directly, instead of changing it back after?
Thanks for your help,
- -Mike.lifeguard
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
iEYEARECAAYFAkqlsAEACgkQst0AR/DaKHuInACguWD1rG+ZjyKaWkRjCplGs4eh
egsAn1wuZxVQx1PGgUFMLSM05iW9EM+N
=/ppz
-----END PGP SIGNATURE-----