f5efd44cc2
[RigoDaemon] acquire UpdatesNotificationResourceLock in non blocking mode
...
We cannot sleep with the activity mutex held without affecting
parallelism so badly. Otoh, we cannot wleep on UpdatesNotificationResourceLock
without affecting 3rd party clients in a way that we would starve them
for a while. So, try to acquire the file lock in non blocking mode
and if it fails, reschedule a new call as soon as the same lock is
freed (exclusive acquire succeeds). This assumes that if the lock
is held back, we will just re-do the whole trick again and again.
2013-12-26 22:52:12 +01:00
f1a925c220
[solo.commands.install] acquire the notification lock in shared NB mode
...
This should avoid starvation in the scenario outlined in code comments.
2013-12-26 22:52:12 +01:00
ec198d632b
[RigoDaemon] make sure that the serializer is always released
...
Even in case of exceptions.
2013-12-26 22:52:12 +01:00
bccbce4a34
[RigoDaemon] use UpdatesNotificationResourceLock to avoid signal spam
...
If this resource lock is acquired, block until the lock is released
so that we will not keep sending status updates to the connected clients
2013-12-26 22:52:12 +01:00
934a7ce14b
[RigoDaemon] drop sleep()
...
With the new locking strategy, this is no longer necessary
2013-12-26 22:52:12 +01:00
c22f4121f6
[RigoDaemon] use ClientSettings() rather than SystemSettings()[id]
2013-12-12 19:22:42 +01:00
ebf63e6060
[entropy.locks] fix shared/exclusive locking mix due to reentrancy, add tests
2013-12-08 16:15:39 +01:00
358c6cc80c
[entropy.client.repository] drop unlocked_sync(), locking happens externally nowadays
2013-12-08 14:42:23 +01:00
1917771b54
[entropy.locks] move Entropy Resources Lock code to a separate module
2013-12-08 14:37:57 +01:00
c0776b07bd
[RigoDaemon] import the threading module directly
2013-12-08 06:17:15 +01:00
0f32ae2696
[RigoDaemon] use systemd-logind shutdown/idle Inhibit() API
...
Inhibit shutdown or idling (see systemd documentation) during
application management or repository update phases.
2013-12-08 06:13:38 +01:00
d8ee468768
[RigoDaemon] add execute_mainloop() wrapper method
2013-12-08 06:13:11 +01:00
e213a5e775
[RigoDaemon] fix regression in api(), the keyword argument must be "sender"
2013-12-07 20:30:07 +01:00
4e597b8769
[RigoDaemon] use a shared lock with clean_package_cache
2013-12-07 20:30:07 +01:00
529203ce5a
[RigoDaemon] maybe_enqueue_kernel_actions: use simulate
2013-12-07 20:30:06 +01:00
4f9d8ef61d
[RigoDaemon] fixup pylint warnings
2013-12-07 20:30:06 +01:00
94e07648ec
[RigoDaemon] add concurrency support \o/
2013-12-07 20:30:06 +01:00
d88e33c40c
[RigoDaemon] use context manager with ReadersWritersSemaphore
2013-12-07 20:30:06 +01:00
82b85704ce
[entropy.output] make TextInterface methods classmethods
...
This makes possible to fix a circular dependency in the Entropy
Resources Lock management methods
2013-12-05 21:36:30 +01:00
7fc8ad14ad
[RigoDaemon] add debug output on _authorize callback method
2013-11-29 11:57:57 +01:00
04048b8818
[RigoDaemon] signal preserved libraries through preserved_libraries_available()
2013-11-29 10:48:50 +01:00
41e0cf576b
[RigoDaemon] migrate to the new PackageActionFactory API
2013-11-23 14:14:48 +01:00
8c5a499e58
[RigoDaemon] set the default terminal (${TERM}) if unset
...
This prevents portage phases printing weird things about missing ${TERM}
2013-11-21 21:48:07 +01:00
3431755aaf
[RigoDaemon] do not trigger kernel-switcher if kernel is already installed
2013-11-13 15:36:32 +01:00
42ab0c38a1
[RigoDaemon] use KernelSwitcher.list() rather than reimplementing the same logic
2013-11-13 13:17:14 +01:00
e93eb24728
[RigoDaemon] implement transparent kernel-switcher support
2013-11-13 13:14:17 +01:00
a323563b22
[RigoDaemon] add "system_restart_needed" signaling
2013-11-05 11:17:32 +01:00
b71ede2b25
[RigoDaemon] implement support for fetch_abort_function
...
This makes possible to immediately interrupt the download process
2013-11-03 15:39:43 +01:00
a0f684b8ae
[RigoDaemon] fix MultipleUrlFetcher stats computation
...
Also introduce stale data handling.
2013-11-03 14:41:21 +01:00
b1642ac2cb
[RigoDaemon] *UrlFetcher: move class variables inside instance
2013-11-03 14:11:22 +01:00
898a200479
[RigoDaemon] always avoid transfer_output() signal floods
2013-11-03 13:59:18 +01:00
60febda4f0
[RigoDaemon] add initial One Click Update notification support
2013-11-03 12:57:08 +01:00
5b9a5ee273
[RigoDaemon] extend updates_available() signal with "One Click Update" information
2013-11-02 14:17:33 +01:00
e02c8da563
[RigoDaemon] fix xml indentation
2013-11-02 14:12:22 +01:00
79261d5086
[RigoDaemon] increase auto_repositories_update randomization space
2013-10-21 13:08:29 +02:00
27659c6dc6
[RigoDaemon] drop access() usage
2013-09-18 08:53:44 +02:00
7ff15f22eb
[RigoDaemon] randomize _auto_repositories_update to avoid bursts of requests
2013-06-10 11:07:18 +02:00
d6295b3e00
[entropy*] create const_mkstemp() as tempfile.mkstemp() wrapper.
...
using /tmp as TMPDIR is a no go, since on modern systems, /tmp is on
tmpfs with a very limited amount of fs size assigned. Use /var/tmp/entropy
(or /var/tmp as fallback) instead.
2013-03-29 12:09:17 +00:00
73abfc6fa0
[RigoDaemon] schedule a reload every 24 hours to cut on Python leaks.
2013-03-04 13:59:44 +00:00
7771d641e2
[entropy.client] extend Client.calculate_updates() to return critical updates info
...
calculate_updates() now returns a dictionary containing the same pieces of
information available before. In addition, it now also returns if critical
updates were found and returned by the method itself.
This is particularly useful for enforcing relaxed deps calculation in case
of critical updates (which is what happens now).
This commit breaks the Entropy Client API.
2013-02-24 13:30:28 +00:00
1655cb3baa
[RigoDaemon] drop send_requested_reply from dbus config, causes deadlock with gdm-3.6
2013-02-12 14:22:56 +01:00
f32ddb95e5
[RigoDaemon] drop any reference to etpUi['debug']
2012-11-06 20:50:47 +01:00
280a7067b3
[entropy.cli] drop module, move code to client/ and server/
...
Even though it is a small code duplication, entropy.cli does make
much less sense.
2012-11-04 21:11:56 +01:00
d9bfa7f11f
[RigoDaemon] half the timer frequency and add some randomization in _auto_repositories_update()
...
When sabayon-weekly is updated, the web service is hit by a huge amount of
requests. This commits add a bit more entropy on the execution of
_auto_repositories_update() (random between 30mins to 2 hours) and reduces
the timer frequency to 8hrs (from 4).
2012-09-08 20:11:22 +02:00
796e71e9ac
[repo] migrate shebangs to /usr/bin/python
2012-08-18 16:53:32 +02:00
d56cd777c8
[RigoDaemon] some minor Python 3.x related fixes
2012-08-18 09:53:18 +02:00
96976374fe
[RigoDaemon] rewrite Installed Repository I/O event handler
...
Rewrite the serialization code of the I/O events coming in when
the Installed Packages Repository is modified (at filesystem level)
to better deal with bursts of events.
The new code uses a "baton" Semaphore as mutex that can be passed
through threads. The MainThread event handler function tries to
acquire the Semaphore in NB mode, if it does, it spawns a thread
that executes all the operations (acquire locks in blocking mode,
calculate updates, etc) and releases the Semaphore once done.
Olympic win!
2012-08-17 12:27:55 +02:00
c5cf96ace0
[RigoDaemon] delay updates calculation by 20 seconds after locks are released
2012-08-17 11:03:58 +02:00
89b114f005
[Rigo*] implement mirrors optimization feature, see bug #3438
2012-08-03 16:26:52 +02:00
107e55a029
[RigoDaemon] fix polkit "Manage Configuration" message field (gnome-polkit guys are idiot!)
2012-08-03 16:17:21 +02:00
Fabio Erculiani