libayatana-indicator/libindicator, branch 0.3.2-0ubuntu1 Ayatana Indicators Shared Library https://openid.arctica-project.org/libayatana-indicator/atom?h=0.3.2-0ubuntu1 2010-01-29T05:39:29+00:00 Adding a debug message for restarting the service. 2010-01-29T05:39:29+00:00 Ted Gould ted@gould.cx 2010-01-29T05:39:29+00:00 urn:sha1:7c5818518d3bd0500c28ea42f0226fa40f066a94 Error handling when building our proxies. 2010-01-29T05:31:45+00:00 Ted Gould ted@gould.cx 2010-01-29T05:31:45+00:00 urn:sha1:1abf840d1c71c3aca0e4ec19a6c5c65a10da0413 Changing timeout to be approximatedly 1 second. 2010-01-29T05:23:18+00:00 Ted Gould ted@gould.cx 2010-01-29T05:23:18+00:00 urn:sha1:52e1a43bd957bac1020d58134ab659553af7c4eb checking the error field for name callback and printing an error when it fails. 2010-01-29T05:07:54+00:00 Ted Gould ted@gould.cx 2010-01-29T05:07:54+00:00 urn:sha1:21eca67f89870abd114f9ccc3ae04e7fd53189b1 In the timeout function mention that we're shutting down, and have an environment variable to stop that. 2010-01-29T05:03:42+00:00 Ted Gould ted@gould.cx 2010-01-29T05:03:42+00:00 urn:sha1:138321d54c11376cabd40d8fb211941508ad5778 Adding in tracking of the restart idle function and making sure we don't do it twice. 2010-01-21T16:30:52+00:00 Ted Gould ted@gould.cx 2010-01-21T16:30:52+00:00 urn:sha1:0219e0e7d51086cc691367e8deda356ac2b6aabc In case we're restarting because of the 'Watch' returning failure we'd have a valid 'service_proxy' object to kill 2010-01-21T16:25:27+00:00 Ted Gould ted@gould.cx 2010-01-21T16:25:27+00:00 urn:sha1:4a7af569bd96c25cbf028fd7144fd2f5d6262a3c Setting up the watch_cb function so that in errors it'll try to restart. As most errors will be fixed by trying again. 2010-01-21T15:49:02+00:00 Ted Gould ted@gould.cx 2010-01-21T15:49:02+00:00 urn:sha1:c72ff4f03fb1fb7c28c2c8c5dda429dc12f45cd5 Update from trunk. 2010-01-18T20:04:38+00:00 Ted Gould ted@gould.cx 2010-01-18T20:04:38+00:00 urn:sha1:a1032ca3e9954f40451c1d0cddcf5e1dcbedd721 It's a fundamental mistake to believe that we can protect people using 2010-01-16T04:28:13+00:00 Ted Gould ted@gould.cx 2010-01-16T04:28:13+00:00 urn:sha1:079c2b517deba95385ac47eea1fd6560aef3aacb this interface from the disconnection. We have no information to say that the new service starting will come up in the same state as the one before it. We need the individual implementers to verify that. Now we need to fix that. This commit does so.