May 2014 - calendarserver-users - lists.macosforge.org

FreeBSD port of CalendarServer 5.1 finished
by Axel Rau 23 Oct '14

23 Oct '14

I just finished a FreeBSD port of CalendarServer 5.1. It is currently waiting for commit to get included in the ports tree: http://www.freebsd.org/cgi/query-pr.cgi?pr=186201 I tested it on FreeBSD 9.2 and did a successful upgrade of the database schema from CalendarServer 4.3 to 5.1. Please test: https://github.com/mc3/fbsd_local_ports Any feedback welcome, Axel --- PGP-Key:29E99DD6 ☀ +49 151 2300 9283 ☀ computing @ chaos claudius

3 4

shared addressbook
by Bodo 25 May '14

25 May '14

Hello calendarserver community, I'am completely new to calendarserver and try to get calendarserver (3.2) up and running on Debian (7.5). I would like to configure an shared address book. If I figured it out correctly address books can only be assigned to users and not to groups. Therefore I set up an user (user1) and created a few contacts. Now I would like to share this address book with other users. Thus I created a group with all users. It looks like calendarserver_manage_principals --add-write-proxy has no effect on this. For calendar's, on the other hand, it works great. Am I doing something wrong? Is there a solution for this except everybody uses user1's login and password? thanks in advance bodo

1 0

Re: [CalendarServer-users] calendarserver-users Digest, Vol 93, Issue 3
by Andy Knasinski 12 May '14

12 May '14

version = "6.0b.trunk-r13264" this is a new install. our previous install went bad due to a python update on this 10.7.4 workstation. i changed debug to error for delegation. > >> On May 12, 2014, at 8:43 AM, Andy Knasinski <ajk(a)nrgsoft.com> wrote: >> >> Not sure how to check on version. > > {9} andre@foci [cs/trunk] % cat calendarserver/version.py > version = "6.0b.trunk-r13468" > >> http://server.com:8008/.calendarserver_version >> >> returns a "2" > > Yeah that?s ? very old, and was only relevant a long time ago (or for very VERY long-range upgrades). > >> proxydbservice is using sqlite > > Have you tried using sqlite3 tools to export / dump the proxydb on a periodic basis, to check for any unexpected changes? > >> i can't seem to find an error log, if i look at web admin >> >> http://server.com:8008/admin/logs/ >> >> i see an access log and server log > > You?re definitely using a recent version; the web admin log viewer is pretty new. This view of error.log should be the same as the on-disk view. > >> We don't run any backup on this box. > > I think the next task is to find the logs. The log location is likely decided by how you installed Calendar Server. If you?re running from an SVN checkout, you?ll find them at data/Logs (relative to the SVN checkout directory). If not, check Calendar Server?s caldavd.plist for ?LogRoot?. Any config keys not defined in caldavd.plist inherit default values from twistedcaldav/stdconfig.py. > > If the error.log doesn?t tell you anything useful at the default log level, you can increase the verbosity of the logs by setting DefaultLogLevel in caldavd.plist to something like info or debug. > > At debug log level, in trunk and recent versions, something like the following is logged at ?debug? level whenever a client changes their delegate configuration: > > 2014-05-12 10:06:00-0700 [DirectoryProxyAMPProtocol,0,] [txdav.who.delegates#debug] Setting delegate assignments for 10000000-0000-0000-0000-000000000001 (read) to [u'10000000-0000-0000-0000-000000000004'] > > You can set the verbosity for specific types of log messages (overriding the default) as follows, allowing you to fine-tune which messages you?d like to see. > > 1) Look at sample log message of the type you wish to become more verbose, in the last square bracketed section before the actual message. In the above case, txdav.who.delegates#debug. This shows the log component and log level of the message. > > 2) Set verbosity for that component in the LogLevels dict in caldavd.plist, e.g. > >  > <key>LogLevels</key> > <dict> > <key>txdav.who.delegates</key> > <string>debug</string> > </dict> > > The proxy / delegation related log messages you might expect to see will depend on which version of the server you?re running. Big changes lately in directory land :)

2 1

Re: [CalendarServer-users] calendarserver-users Digest, Vol 92, Issue 5
by Andy Knasinski 12 May '14

12 May '14

Not sure how to check on version. http://server.com:8008/.calendarserver_version returns a "2" proxydbservice is using sqlite i can't seem to find an error log, if i look at web admin http://server.com:8008/admin/logs/ i see an access log and server log We don't run any backup on this box. On Apr 21, 2014, at 2:00 PM, calendarserver-users-request(a)lists.macosforge.org wrote: > What version of Calendar Server are you using? Is the server configuration customized in any way with respect to the proxy (delegation) data storage? This is represented in the ProxyDBService dict in caldavd.plist. It?s either going to be storing this data in an sqlite DB, or in a postgres DB. Sqlite is the default. > > Also, check Calendar Server?s error.log, which might be at /var/log/caldavd/error.log. > > Do you have any sort of backup automation that might be causing this? I?m not intimately familiar with our usage of the sqlite DB for proxy storage, but it?s sqlite3, it could in theory become damaged in an improper shutdown. Note also that any backups of this file should only be taken when the service is offline, to ensure that the sqlite DB file is consistent and quiesced.

2 1

Questions regarding CardDAV server VCard "Categories" and OSX Contacts "Groups"
by tewarfel 02 May '14

02 May '14

I've been experimenting with setting up a home instance of Calendar Server to sync Contacts and Calendar data across our devices, trying both the 5.1 and 5.2 builds running under OSX 10.8.5, and built using Homebrew'ed Python 2.7.6 --with-brewed-OpenSSL. I'm using the XML File Directory Service + accounts.xml to create a single user for now. I'm trying to get multiple devices to all use a shared common set of calendars and contacts. The Calendar Server component works as expected - I can create multiple different calendars on one device, and they all show up (as expected) on the other devices as distinct named calendars on teh other devices. Contacts are NOT working as expected, and I'm not sure if I have a faulty understanding of how they are supposed to work (or if I'm less clever than I thought I was as far as getting this to build and run). Using Contacts.app from OSX 10.8.5 on a Macbook, I can add an account for the CardDAV server (separately from the CalDAV server), and can set that server account to be my primary account within Contacts. In addition, I deleted all the local contacts off the laptop to minimize confusion. I close and restart Contacts, and see the server listed without any contacts. I can then create "groups" under that server within Contacts (by selecting File -> New Group), but when I try to import a CardDAV file, everything just goes into the "All Contacts" heading on the server. No matter. I select "Last Import", "Select All", and drag the entries to my newly-created Contacts group ("group" within Contacts, "Category" in Vcard parlance). I repeat this until I've built up 3-4 different contact groups in Contacts and they seem to work correctly on the laptop. Now the confusing part.... I take a second OSX machine (tried with both OSX 10.8 and 10.9) and point it at the same server with the same account. All I get is one "All Contacts" heading for the server. All of the contact entries are in there, but there is no separation by the different Categories (Contact groups). Same if I point an iOS 6.1 handheld device at the server. And if I just point my web browser at the server (and authenticate with the info in accounts.xml), I see all my vcard entries in https://192.168.1.114:8443/addressbooks/users/{insert-username-here}/addres… but there are no other address books other than "addressbook" listed. I reboot the original laptop and restart contacts... and the group separations are still present. I delete the CardDAV account settings on the laptop, restart Contacts, and (as expected) everything is gone. Re-add the account settings, and <poof> all the Contacts are back AND they are still in the original separate groups I had created. (I'm assuming the group separation metadata somehow persisted across the account deletion and recreation). Separate computer, same OS, all one blob with no groups. Same computer, but delete account settings and then re-add them -> the separate groups (Categories) are still there. Unlike the multiple Calendars (where they are kept separate and imported separately) the Contact separation into groups appears confined to the original laptop - all other machines connecting to the CardDAV server just see the big blob of All Contacts. Is my expectation of multiple Contacts Groups, once created by one laptop using the CardDAV server being available on other laptops (just like multiple calendars are) incorrect, or is this a (hopefully well-known) (mis)configuration error? Or, do I need to do something to explicitly pre-create the contact groups in PostgreSQL (to represent the VCard categories) prior to importing the actual contact data to them on the laptop? Comments / suggestions / redirections appreciated. Thanks, -Tom -- View this message in context: http://mac-os-forge.2317878.n4.nabble.com/Questions-regarding-CardDAV-serve… Sent from the Calendar Server - Users mailing list archive at Nabble.com.

1 0

Upgrading schema
by Pascal Dallaire 01 May '14

01 May '14

Hi to everyone on the list, I’m trying to migrate database+attachments from a CalendarServer 4.2 to 5.2. I used the calendarserver_backup script and was able to get a tgz… I used the same script from 5.2 to restore, there were some errors during the restore, next time I open the server (I used the FailIfUpgradeNeeded=False so that it would auto-upgrade automatically). When I run the server next time, it tries to apply the upgrade the schema from 12 to 13. Then I get a lot of errors, Traceback, with the last line being : --------- pg.ProgrammingError: ERROR: relation "attachment_id_seq" already exists --------- and then : --------- 2014-05-01 05:40:31-0400 [-] [txdav.base.datastore.subpostgres#warn] received postgres stdout 'waiting for server to shut down....' 2014-05-01 05:40:32-0400 [-] [txdav.base.datastore.subpostgres#warn] received postgres stdout ' done\nserver stopped\n' 2014-05-01 05:40:32-0400 [-] [txdav.base.datastore.subpostgres#warn] postgres process ended with status 0 2014-05-01 05:40:32-0400 [-] [twisted.internet.base#info] Main loop terminated. 2014-05-01 05:40:32-0400 [-] [twisted.application.app#info] Server Shut Down. --------- Is it possible for someone who knows to nudge me toward a way to repair these errors? Thanks in advance Pascal

2 1