3:52 a.m.
Hello,
I have one small problem: I cannot select which local Evolution calendar
SyncEvolution actually talks to.
Apparently it uses the first entry in
the /apps/evolution/calendar/sources gconf entry that it can find. Since
the order of these entries isn't changeable and I need to separate
"work" from "home" calendar entries (with separate mobiles to sync to
…), I'd like some way to change that.
Either this is not possible yet, or I'm blind and didn't find it in the
documentation …?
Otherwise, thanks for the excellent tool; it works with our Oracle
calendar like a charm (using the values in the Oracle branch).
-- Matthias
4:52 a.m.
On Tue, 2009-11-17 at 11:52 +0000, Matthias Urlichs wrote:
Apparently it uses the first entry in
the /apps/evolution/calendar/sources gconf entry that it can find. Since
the order of these entries isn't changeable and I need to separate
"work" from "home" calendar entries (with separate mobiles to sync
to
…), I'd like some way to change that.
Either this is not possible yet, or I'm blind and didn't find it in the
documentation …?
The README points towards the description of the source properties
compiled into the code; here's the relevant one:
$ syncevolution --source-property evolutionsource=?
'--source-property evolutionsource=?'
Picks one of backend data sources:
enter either the name or the full URL.
Most backends have a default data source,
like for example the system address book.
Not setting this property selects that default
data source.
To get a full list of available data sources,
run syncevolution without parameters. The name
is printed in front of the colon, followed by
the URL. Usually the name is unique and can be
used to reference the data source. The default
data source is marked with <default> after the
URL, if there is a default.
Otherwise, thanks for the excellent tool; it works with our Oracle
calendar like a charm (using the values in the Oracle branch).
Hmm, then you've been more successful than myself so far. At the moment,
the Oracle interoperability server rejects all iCalendar 2.0 events sent
by us, for unknown reasons.
--
Best Regards, Patrick Ohly
The content of this message is my personal opinion only and although
I am an employee of Intel, the statements I make here in no way
represent Intel's position on the issue, nor am I authorized to speak
on behalf of Intel on this matter.
5:07 a.m.
New subject: [SyncEvolution] Oracle interoperability
On Tue, 2009-11-17 at 13:52 +0100, Patrick Ohly wrote:
> Otherwise, thanks for the excellent tool; it works with our
Oracle
> calendar like a charm (using the values in the Oracle branch).
Hmm, then you've been more successful than myself so far. At the
moment, the Oracle interoperability server rejects all iCalendar 2.0
events sent by us, for unknown reasons.
Good for me, then.
NB: I used this server URL, which seems to be somewhat standard:
https://calserver.company.example/ocst-bin/ocas.fcgi
3:23 p.m.
Matthias,
What is meant by "Oracle branch"?
<quote>
Otherwise, thanks for the excellent tool; it works with our Oracle
calendar like a charm (using the values in the Oracle branch).
</quote>
I, too, am trying to get syncevolution to work with an Oracle Calendar
OCAS SyncML server (I am the administrator of that system, btw). Right
now the *default* Ubuntu Lucid 10.04 install of syncevolution+Genesis
appears to be configured correctly. The logs show a successful
connection to the SyncML server (so we know the URL/credentials are
correct), but the sync concludes with a 503 error (have removed
indenting for legibility):
<quote>
[2010-06-16 15:51:47.622] 'processStatus' - Processing incoming
Status [--][++] [->end] [->enclosing]
o [2010-06-16 15:51:47.622] Started processing Command 'Status'
(incoming MsgID=1, CmdID=6)
o [2010-06-16 15:51:47.622] WARNING: RECEIVED NON-OK STATUS 503 for
command 'Alert' (outgoing MsgID=1, CmdID=4)
o [2010-06-16 15:51:47.622] - TargetRef (remoteID) = './Contacts'
o [2010-06-16 15:51:47.622] - SourceRef (localID) = './addressbook'
o [2010-06-16 15:51:47.622] - Item data =
o [2010-06-16 15:51:47.622] Found matching command 'Alert' for Status
</quote>
And ...
<quote>
[2010-06-16 15:51:47.623] 'SessionAbort' - Aborting Session,
Status=503, ProblemSource=REMOTE [--][++] [->end] [->enclosing]
@ [2010-06-16 15:51:47.623] WARNING: Aborting Session with Reason
Status 503 (REMOTE problem) ***
@ [2010-06-16 15:51:47.623] --------------- Ignoring all commands
in this message (after 0 sec. request processing, 0 sec. total)
with Status 514 (0=none) from here on
–[2010-06-16 15:51:47.623] End of 'SessionAbort' [->top] [->enclosing]
</quote>
I can confirm that the syncserver is working (I can sync with both a
Blackberry using Nexthaus' SyncJe and a Palm Treo650 using Synthesis'
Sync client).
For the syncevolution folks. As I am the server admin I have complete
access to the server logs and configuration and I am quite willing to
assist in any/all troubleshooting to help get this reliably working. I
think it would be very helpful for our non-Windows/MacOS users. I see in
the list archives two threads about Oracle Calendar syncing from Oct/Nov
2009 but neither appears to have posts that indicate full resolution of
the issues they mention.
Again, this is an Oracle Calendar Server v10.1.2.3 (not Oracle Beehive).
Best regards,
Guy
On 11/17/2009 05:52 AM, Matthias Urlichs wrote:
Hello,
I have one small problem: I cannot select which local Evolution calendar
SyncEvolution actually talks to.
Apparently it uses the first entry in
the /apps/evolution/calendar/sources gconf entry that it can find. Since
the order of these entries isn't changeable and I need to separate
"work" from "home" calendar entries (with separate mobiles to sync
to
…), I'd like some way to change that.
Either this is not possible yet, or I'm blind and didn't find it in the
documentation …?
Otherwise, thanks for the excellent tool; it works with our Oracle
calendar like a charm (using the values in the Oracle branch).
-- Matthias
_______________________________________________
SyncEvolution mailing list
SyncEvolution(a)syncevolution.org
http://lists.syncevolution.org/listinfo/syncevolution
11:54 p.m.
On Wed, 2010-06-16 at 23:23 +0100, Guy Stalnaker wrote:
What is meant by "Oracle branch"?
At one point there was a branch in our source code repository for
Oracle. It had a new template for Oracle. This has been rolled into the
main branch and is in the latest releases, so you can use "--configure
--sync-property syncurl=<my url> ... oracle".
I, too, am trying to get syncevolution to work with an Oracle
Calendar
OCAS SyncML server (I am the administrator of that system, btw). Right
now the *default* Ubuntu Lucid 10.04 install of syncevolution+Genesis
appears to be configured correctly. The logs show a successful
connection to the SyncML server (so we know the URL/credentials are
correct), but the sync concludes with a 503 error (have removed
indenting for legibility):
<quote>
[2010-06-16 15:51:47.622] 'processStatus' - Processing incoming
Status [--][++] [->end] [->enclosing]
o [2010-06-16 15:51:47.622] Started processing Command 'Status'
(incoming MsgID=1, CmdID=6)
o [2010-06-16 15:51:47.622] WARNING: RECEIVED NON-OK STATUS 503 for
command 'Alert' (outgoing MsgID=1, CmdID=4)
o [2010-06-16 15:51:47.622] - TargetRef (remoteID) = './Contacts'
o [2010-06-16 15:51:47.622] - SourceRef (localID) = './addressbook'
o [2010-06-16 15:51:47.622] - Item data =
o [2010-06-16 15:51:47.622] Found matching command 'Alert' for Status
</quote>
./Contacts is correct and matches the Oracle template (ignoring case,
which IMHO doesn't matter).
Perhaps this article here helps? It is about Behive, but I can imagine
that Calendar Server also needs server-side configuration changes.
http://syncevolution.org/wiki/nokia-n900-oracle-beehive
--
Best Regards, Patrick Ohly
The content of this message is my personal opinion only and although
I am an employee of Intel, the statements I make here in no way
represent Intel's position on the issue, nor am I authorized to speak
on behalf of Intel on this matter.
1:18 p.m.
Patrick,
On 06/17/2010 01:54 AM, Patrick Ohly wrote:
> On Wed, 2010-06-16 at 23:23 +0100, Guy Stalnaker wrote:
>> What is meant by "Oracle branch"?
>
> At one point there was a branch in our source code repository for
> Oracle. It had a new template for Oracle. This has been rolled into the
> main branch and is in the latest releases, so you can use "--configure
> --sync-property syncurl=<my url> ... oracle".
Thank you very much for your reply. The log of my latest attempt is
attached. Here is the situation.
Server: Oracle Calendar OCAS v10.1.2.3.4
Client: SyncEvolution 1.0beta2a
Host: Ubuntu Lucid 10.04 32-bit
I have both manually created a profile and (following your explanation)
I've used the built-in oracle template. Both result in the exact same
503 error. As you can see from the log, credential authentication is
successful. As is the query for the Sync server info:
<quote>
[2010-06-17 14:07:45.819] 'DevInf_Analyze' - Analyzing remote
devInf [--][++] [->end] [->enclosing]
# [2010-06-17 14:07:45.820] Device ID='OracleSyncServer', Type='server',
Model=''
# [2010-06-17 14:07:45.820] Manufacturer='Oracle', OEM='Oracle'
# [2010-06-17 14:07:45.820] Softwarevers='10.1.2',
Firmwarevers='10.1.2', Hardwarevers='10.1.2'
# [2010-06-17 14:07:45.820] SyncML Version: SyncML/1.2
# [2010-06-17 14:07:45.820] SyncML capability flags: wantsNOC=Yes,
canHandleUTC=Yes, supportsLargeObjs=Yes
# [2010-06-17 14:07:45.820] Detected Oracle OCS Server - suppress
dynamic X-nnnn TYPE params, and <filter>
# [2010-06-17 14:07:45.820] OCS with device that
has not guaranteed
unique ID - use user+devid hash for Source LocURI
</quote>
SyncEvolution both successfully authenticates a session and it
successfully queries and parses the response for devInf. The failure
comes next.
<quote>
[2010-06-17 14:07:45.827] 'issue' - issuing command,
Cmd=Status [--][++] [->end] [->enclosing]
# [2010-06-17 14:07:45.827] Status Code 200 issued for Cmd=Results,
(incoming MsgID=1, CmdID=4)
# [2010-06-17 14:07:45.827] - SourceRef (remoteID) = './devinf12'
# [2010-06-17 14:07:45.827] Status: issued as (outgoing MsgID=2,
CmdID=2), not waiting for status
# [2010-06-17 14:07:45.827] Deleted command 'Status' (outgoing MsgID=2,
CmdID=2)
# [2010-06-17 14:07:45.827] Outgoing Message size is now 384 bytes
–[2010-06-17 14:07:45.827] End of 'issue' [->top] [->enclosing]
o [2010-06-17 14:07:45.828] Deleted command 'Results' (incoming
MsgID=1, CmdID=4)
–[2010-06-17 14:07:45.828] End of 'processCmd' [->top] [->enclosing]
+ [2010-06-17 14:07:45.828] Created command 'Status' (incoming)
+ +
– [2010-06-17 14:07:45.828] 'processStatus' - Processing incoming
Status [--][++] [->end] [->enclosing]
o [2010-06-17 14:07:45.828] Started processing Command 'Status'
(incoming MsgID=1, CmdID=5)
o [2010-06-17 14:07:45.828] WARNING: RECEIVED NON-OK STATUS 503 for
command 'Alert' (outgoing MsgID=1, CmdID=3)
o [2010-06-17 14:07:45.828] - TargetRef (remoteID) = './calendar/tasks'
o [2010-06-17 14:07:45.828] - SourceRef (localID) = './todo'
o [2010-06-17 14:07:45.828] - Item data =
o [2010-06-17 14:07:45.828] Found matching command 'Alert' for Status -
Synthesis SyncML Engine 3.4.0.5 Log (p15 of 19)
o +
</quote>
The abort happens soon after.
You'll know why this "Command 'Alert'" is being sent to the server?
What
is the expected server response?
Using Oracle-supplied steps I created a sync profile for SyncEvolution
(much like the article about SE and Beehive mentions). I can see that
the Sync server records the connection in data for my user account, but
that data is minimal as the sync session is terminated after the 503
error (I've included the full directory listing for my data so you can
see what the various files look like on the Oracle side of things; the
364A7A16 set are from use of a Palm Synthesis; the IMEI is a Blackberry
Nexthaus client; the NL-283134 is NotifyLink Enterprise Server):
<quote>
[2:06pm nak@wisccal-sync] ~> sudo ls -l
ocas10g/linkdb/calserv/jstalnak,user=/
wisccal-sync Password:
total 678
-rw-r----- 1 oracle dba 19034 Jun 5 13:03
364A7A16__calendar_events.db
-rw-r----- 1 oracle dba 85 Mar 30 17:07
364A7A16__calendar_tasks.db
-rw-r----- 1 oracle dba 75 Mar 30 17:07 364A7A16__contacts.db
-rw-r----- 1 oracle dba 51308 Jun 16 17:24
IMEI_3193ec46__calendar_events.db
-rw-rw---- 1 oracle dba 771 Jun 18 2009
IMEI_3193ec46__calendar_tasks.db
-rw-rw---- 1 oracle dba 914 Sep 2 2009
IMEI_3193ec46__contacts.db
-rw-r----- 1 oracle dba 20070 Nov 2 2009
IMEI_32caae79__calendar_events.db
-rw-r----- 1 oracle dba 787 Nov 8 2009
IMEI_32caae79__calendar_tasks.db
-rw-r----- 1 oracle dba 41480 Nov 2 2009
IMEI_32caae79__contacts.db
-rw-rw---- 1 oracle dba 4795 Jun 17 14:06 index.idx
-rw-r----- 1 oracle dba 76899 Jan 8 17:07
NL-283134__calendar_events.db
-rw-r----- 1 oracle dba 1452 Jan 8 17:07
NL-283134__calendar_tasks.db
-rw-r----- 1 oracle dba 98567 Mar 31 18:43
NL-289054__calendar_events.db
-rw-r----- 1 oracle dba 1452 Mar 31 18:43
NL-289054__calendar_tasks.db
-rw-r----- 1 oracle dba 83 Jun 16 09:28
syncevolution-32cbd467-c9a4-437a-ab8e-1718ab773846__calendar_events.db
-rw-r----- 1 oracle dba 77 Jun 16 09:28
syncevolution-32cbd467-c9a4-437a-ab8e-1718ab773846__calendar_tasks.db
-rw-r----- 1 oracle dba 79 Jun 16 09:28
syncevolution-32cbd467-c9a4-437a-ab8e-1718ab773846__contacts.db
-rw-r----- 1 oracle dba 83 Jun 17 14:06
syncevolution-3c7dc774-60b1-4255-9ff2-69f3395866cc__calendar_events.db
-rw-r----- 1 oracle dba 77 Jun 17 14:06
syncevolution-3c7dc774-60b1-4255-9ff2-69f3395866cc__calendar_tasks.db
-rw-r----- 1 oracle dba 79 Jun 17 14:06
syncevolution-3c7dc774-60b1-4255-9ff2-69f3395866cc__contacts.db
</quote>
It's obvious from the file sizes alone that the syncevolution data files
have minimal information as all are less than 90 *bytes*.
How can we trouble-shoot this?
Again, I'm the Oracle Calendar admin so I have complete access to all
facets of the Oracle OCAS (sync server) configuration and logs. I can
give you WAY more debug log data than you really want :-)
Where I want to go with my testing is a process that will allow users of
MacOS Apple iCal a method to connect to and read/write their Oracle
Calendar data. Right now that is not possible. We've a project to
replace our Oracle Calendar but that's going to take two years. If I can
get our many Mac users a method to access their Oracle account using
Apple iCal, I'd be a very happy person. Because syncevolution writes
its data to a file using icalendar, there's a chance that same data file
may work with Apple iCal (as it works for Evolution) since Apple iCal
supports iCalendar (!?).
Best regards,
Guy
./Contacts is correct and matches the Oracle template (ignoring case,
which IMHO doesn't matter).
Perhaps this article here helps? It is about Behive, but I can imagine
that Calendar Server also needs server-side configuration changes.
http://syncevolution.org/wiki/nokia-n900-oracle-beehive
12:29 a.m.
Hello Guy,
That 503 error sounds familiar to me. I don't have detail insight into OCS, but
we've done a lot of testing together with the SyncML guys at Oracle. We've seen
that 503 usually when a previous sync with the same user or device has not sucessfully
completed, and a session was still pending somehow. Just leaving the server alone for
5mins and trying again has usually worked.
This situation easily happens when testing new stuff, but in normal operation this is rare
so end users don't run into that too often. Still, some do and we have a FAQ entry on
the Synthesis site for that: http://www.synthesis.ch/faq.php?lang=e (BTW: Altough that FAQ
is for our commercial products, it might be helpful regarding compatibility issues with
libsynthesis based clients as well.
Best Regards,
Lukas Zeller
On Jun 17, 2010, at 22:18 , Guy Stalnaker wrote:
Patrick,
> On 06/17/2010 01:54 AM, Patrick Ohly wrote:
>> On Wed, 2010-06-16 at 23:23 +0100, Guy Stalnaker wrote:
>>> What is meant by "Oracle branch"?
>>
>> At one point there was a branch in our source code repository for
>> Oracle. It had a new template for Oracle. This has been rolled into the
>> main branch and is in the latest releases, so you can use "--configure
>> --sync-property syncurl=<my url> ... oracle".
Thank you very much for your reply. The log of my latest attempt is attached. Here is
the situation.
Server: Oracle Calendar OCAS v10.1.2.3.4
Client: SyncEvolution 1.0beta2a
Host: Ubuntu Lucid 10.04 32-bit
I have both manually created a profile and (following your explanation) I've used the
built-in oracle template. Both result in the exact same 503 error. As you can see from
the log, credential authentication is successful. As is the query for the Sync server
info:
<quote>
[2010-06-17 14:07:45.819] 'DevInf_Analyze' - Analyzing remote
devInf [--][++] [->end] [->enclosing]
# [2010-06-17 14:07:45.820] Device ID='OracleSyncServer', Type='server',
Model=''
# [2010-06-17 14:07:45.820] Manufacturer='Oracle', OEM='Oracle'
# [2010-06-17 14:07:45.820] Softwarevers='10.1.2',
Firmwarevers='10.1.2', Hardwarevers='10.1.2'
# [2010-06-17 14:07:45.820] SyncML Version: SyncML/1.2
# [2010-06-17 14:07:45.820] SyncML capability flags: wantsNOC=Yes,
canHandleUTC=Yes, supportsLargeObjs=Yes
# [2010-06-17 14:07:45.820] Detected Oracle OCS Server - suppress
dynamic X-nnnn TYPE params, and <filter>
# [2010-06-17 14:07:45.820] OCS with device that
has not guaranteed
unique ID - use user+devid hash for Source LocURI
</quote>
SyncEvolution both successfully authenticates a session and it successfully queries and
parses the response for devInf. The failure comes next.
<quote>
[2010-06-17 14:07:45.827] 'issue' - issuing command,
Cmd=Status [--][++] [->end] [->enclosing]
# [2010-06-17 14:07:45.827] Status Code 200 issued for Cmd=Results, (incoming MsgID=1,
CmdID=4)
# [2010-06-17 14:07:45.827] - SourceRef (remoteID) = './devinf12'
# [2010-06-17 14:07:45.827] Status: issued as (outgoing MsgID=2, CmdID=2), not waiting
for status
# [2010-06-17 14:07:45.827] Deleted command 'Status' (outgoing MsgID=2, CmdID=2)
# [2010-06-17 14:07:45.827] Outgoing Message size is now 384 bytes
–[2010-06-17 14:07:45.827] End of 'issue' [->top] [->enclosing]
o [2010-06-17 14:07:45.828] Deleted command 'Results' (incoming MsgID=1,
CmdID=4)
–[2010-06-17 14:07:45.828] End of 'processCmd' [->top] [->enclosing]
+ [2010-06-17 14:07:45.828] Created command 'Status' (incoming)
+ +
– [2010-06-17 14:07:45.828] 'processStatus' - Processing incoming Status [--][++]
[->end] [->enclosing]
o [2010-06-17 14:07:45.828] Started processing Command 'Status' (incoming
MsgID=1, CmdID=5)
o [2010-06-17 14:07:45.828] WARNING: RECEIVED NON-OK STATUS 503 for command
'Alert' (outgoing MsgID=1, CmdID=3)
o [2010-06-17 14:07:45.828] - TargetRef (remoteID) = './calendar/tasks'
o [2010-06-17 14:07:45.828] - SourceRef (localID) = './todo'
o [2010-06-17 14:07:45.828] - Item data =
o [2010-06-17 14:07:45.828] Found matching command 'Alert' for Status - Synthesis
SyncML Engine 3.4.0.5 Log (p15 of 19)
o +
</quote>
The abort happens soon after.
You'll know why this "Command 'Alert'" is being sent to the server?
What is the expected server response?
Using Oracle-supplied steps I created a sync profile for SyncEvolution (much like the
article about SE and Beehive mentions). I can see that the Sync server records the
connection in data for my user account, but that data is minimal as the sync session is
terminated after the 503 error (I've included the full directory listing for my data
so you can see what the various files look like on the Oracle side of things; the 364A7A16
set are from use of a Palm Synthesis; the IMEI is a Blackberry Nexthaus client; the
NL-283134 is NotifyLink Enterprise Server):
<quote>
[2:06pm nak@wisccal-sync] ~> sudo ls -l ocas10g/linkdb/calserv/jstalnak,user=/
wisccal-sync Password:
total 678
-rw-r----- 1 oracle dba 19034 Jun 5 13:03 364A7A16__calendar_events.db
-rw-r----- 1 oracle dba 85 Mar 30 17:07 364A7A16__calendar_tasks.db
-rw-r----- 1 oracle dba 75 Mar 30 17:07 364A7A16__contacts.db
-rw-r----- 1 oracle dba 51308 Jun 16 17:24 IMEI_3193ec46__calendar_events.db
-rw-rw---- 1 oracle dba 771 Jun 18 2009 IMEI_3193ec46__calendar_tasks.db
-rw-rw---- 1 oracle dba 914 Sep 2 2009 IMEI_3193ec46__contacts.db
-rw-r----- 1 oracle dba 20070 Nov 2 2009 IMEI_32caae79__calendar_events.db
-rw-r----- 1 oracle dba 787 Nov 8 2009 IMEI_32caae79__calendar_tasks.db
-rw-r----- 1 oracle dba 41480 Nov 2 2009 IMEI_32caae79__contacts.db
-rw-rw---- 1 oracle dba 4795 Jun 17 14:06 index.idx
-rw-r----- 1 oracle dba 76899 Jan 8 17:07 NL-283134__calendar_events.db
-rw-r----- 1 oracle dba 1452 Jan 8 17:07 NL-283134__calendar_tasks.db
-rw-r----- 1 oracle dba 98567 Mar 31 18:43 NL-289054__calendar_events.db
-rw-r----- 1 oracle dba 1452 Mar 31 18:43 NL-289054__calendar_tasks.db
-rw-r----- 1 oracle dba 83 Jun 16 09:28
syncevolution-32cbd467-c9a4-437a-ab8e-1718ab773846__calendar_events.db
-rw-r----- 1 oracle dba 77 Jun 16 09:28
syncevolution-32cbd467-c9a4-437a-ab8e-1718ab773846__calendar_tasks.db
-rw-r----- 1 oracle dba 79 Jun 16 09:28
syncevolution-32cbd467-c9a4-437a-ab8e-1718ab773846__contacts.db
-rw-r----- 1 oracle dba 83 Jun 17 14:06
syncevolution-3c7dc774-60b1-4255-9ff2-69f3395866cc__calendar_events.db
-rw-r----- 1 oracle dba 77 Jun 17 14:06
syncevolution-3c7dc774-60b1-4255-9ff2-69f3395866cc__calendar_tasks.db
-rw-r----- 1 oracle dba 79 Jun 17 14:06
syncevolution-3c7dc774-60b1-4255-9ff2-69f3395866cc__contacts.db
</quote>
It's obvious from the file sizes alone that the syncevolution data files have minimal
information as all are less than 90 *bytes*.
How can we trouble-shoot this?
Again, I'm the Oracle Calendar admin so I have complete access to all facets of the
Oracle OCAS (sync server) configuration and logs. I can give you WAY more debug log data
than you really want :-)
Where I want to go with my testing is a process that will allow users of MacOS Apple iCal
a method to connect to and read/write their Oracle Calendar data. Right now that is not
possible. We've a project to replace our Oracle Calendar but that's going to take
two years. If I can get our many Mac users a method to access their Oracle account using
Apple iCal, I'd be a very happy person. Because syncevolution writes its data to a
file using icalendar, there's a chance that same data file may work with Apple iCal
(as it works for Evolution) since Apple iCal supports iCalendar (!?).
Best regards,
Guy
>
> ./Contacts is correct and matches the Oracle template (ignoring case,
> which IMHO doesn't matter).
>
> Perhaps this article here helps? It is about Behive, but I can imagine
> that Calendar Server also needs server-side configuration changes.
>
> http://syncevolution.org/wiki/nokia-n900-oracle-beehive
>
<syncevolution-log.html>_______________________________________________
SyncEvolution mailing list
SyncEvolution(a)syncevolution.org
http://lists.syncevolution.org/listinfo/syncevolution
Lukas Zeller (luz(a)synthesis.ch)
-
Synthesis AG, SyncML Solutions & Sustainable Software Concepts
info(a)synthesis.ch, http://www.synthesis.ch
2:32 a.m.
Hello Guy!
Lukas already replied to the 503 error. I'm afraid I don't have anything
to add to that. Hmm, or perhaps I do - see below.
On Thu, 2010-06-17 at 21:18 +0100, Guy Stalnaker wrote:
Where I want to go with my testing is a process that will allow users
of
MacOS Apple iCal a method to connect to and read/write their Oracle
Calendar data. Right now that is not possible. We've a project to
replace our Oracle Calendar but that's going to take two years. If I can
get our many Mac users a method to access their Oracle account using
Apple iCal, I'd be a very happy person. Because syncevolution writes
its data to a file using icalendar, there's a chance that same data file
may work with Apple iCal (as it works for Evolution) since Apple iCal
supports iCalendar (!?).
But how would you deploy this? It's true that you end up with all events
in a single iCalendar 2.0 file, but that's in the Linux home directory.
The connection from there to Apple iCal would still be missing.
It would be interesting to look into writing a CalDAV backend for
SyncEvolution. Given that, it should be possible to read/write the data
in a server that can be accessed by Apple iCal (which supports CalDAV).
But isn't CalDAV also supported by Oracle? Just wondering why iCal is
not already supported.
Also note that calendar synchronization with Oracle ran into issues with
iCalendar 2.0 data and the UID/RECURRENCE-ID properties. The main issue
is that Oracle expects the UID to be unique among all SyncML items. But
in iCalendar 2.0, multiple such items may (and must!) have the same UID
when they belong to the same meeting series. I never figured out how to
resolve this, so interoperability testing with Oracle got stuck.
My suggestion would be to get contact syncing working first, to avoid
these issues. Perhaps they caused sync failures and thus the 503 error
after all.
--
Best Regards, Patrick Ohly
The content of this message is my personal opinion only and although
I am an employee of Intel, the statements I make here in no way
represent Intel's position on the issue, nor am I authorized to speak
on behalf of Intel on this matter.
3630
days inactive
3846
days old
syncevolution@synevolution.org
7 comments
4 participants
participants (4)
-
Guy Stalnaker -
Lukas Zeller -
Matthias Urlichs -
Patrick Ohly