search for: workers

A full example with dovecot debug symbols activated (note since I have activate valgrind, no more segfault it?s normal I presume ): Dec 4 12:09:40 dovecot: auth-worker: Error: ==3071== Invalid read of size 8 Dec 4 12:09:40 dovecot: auth-worker: Error: ==3053== Invalid read of size 8 Dec 4 12:09:40 dovecot: auth-worker: Error: ==3057== Invalid read of size 8 Dec 4 12:09:40 dovecot:

On 2 Dec 2018, at 22.22, Guillaume via dovecot <dovecot at dovecot.org> wrote: > > I also have this kind of segfault since the update : > > Dec 2 21:12:11 xxxxxxx dovecot: auth-worker: Error: *** Error in `dovecot/auth': double free or corruption (fasttop): 0x000055573bb99f70 Is this easy to reproduce? Can you try with valgrind? It will slow down the logins a bit though.

Hi ! Yes it happens every 2 to 5 minutes so no problem in reproducing. I made the change with valgrind, I have now this in my log : Dec 4 12:00:27 xxxxx dovecot: auth-worker: Error: ==2436== Invalid read of size 8 Dec 4 12:00:27 xxxxx dovecot: auth-worker: Error: ==2393== Invalid read of size 8 Dec 4 12:00:27 xxxxx dovecot: auth-worker: Error: ==2476== Invalid read of size 8 Dec 4 12:00:27

If userspace does VHOST_VSOCK_SET_GUEST_CID before VHOST_SET_OWNER we can race where: 1. thread0 calls vhost_transport_send_pkt -> vhost_work_queue 2. thread1 does VHOST_SET_OWNER which calls vhost_worker_create. 3. vhost_worker_create will set the dev->worker pointer before setting the worker->vtsk pointer. 4. thread0's vhost_work_queue will see the dev->worker pointer is set and

If userspace does VHOST_VSOCK_SET_GUEST_CID before VHOST_SET_OWNER we can race where: 1. thread0 calls vhost_transport_send_pkt -> vhost_work_queue 2. thread1 does VHOST_SET_OWNER which calls vhost_worker_create. 3. vhost_worker_create will set the dev->worker pointer before setting the worker->vtsk pointer. 4. thread0's vhost_work_queue will see the dev->worker pointer is set and

I have the following script: #!/bin/sh PATH=$PATH:/usr/local/bin #Expects to be run after midnight on the first of the month # to archive all the previous months mail #Date Run: TODAY=`date "+%Y-%m-%d"` #last month in YYYY/MM YEAR_LAST_MONTH=`date -v-1d "+%Y/%m"` #1st of last month as 01-Mon-YYYY FIRST_LAST_MONTH=`date -v-1d "+01-%b-%Y"` echo 'TODAY='

On 5/31/23 10:15 AM, Mike Christie wrote: >>> rcu would work for your case and for what Jason had requested. >> Yeah, so you already have some patches? >> >> Do you want to send it to solve this problem? >> > Yeah, I'll break them out and send them later today when I can retest > rebased patches. > Just one question. Do you core vhost developers

I have been facing this damn ugly issue since quite sometime. So this worker reads RSS feed from Yahoo news and stores it in a DB. The worker is scheduled to run at every 40 minutes and my background_schedules.yml looks like this: feed_worker: :class: :feed_worker :job_key: :feed_worker_key :worker_method: :do_work :trigger_args: :start: <%= Time.now + 5 %>

...=randrw --bs=4k \ --ioengine=libaio --iodepth=128 --numjobs=3 the single vhost worker thread will become a bottlneck and we are stuck at around 500K IOPs no matter how many jobs, virtqueues, and CPUs are used. To better utilize virtqueues and available CPUs, this patch allows userspace to create workers and bind them to vqs. You can have N workers per dev and also share N workers with M vqs. With the patches and doing a worker per vq, we can scale to at least 16 vCPUs/vqs (that's my system limit) with the same command fio command above with numjobs=16: fio --filename=/dev/sdb --direct=1 --r...

...test&greatest backgroundrb with rails 2.0.2 on ubuntu dapper. I''ve made an app that lets you run a query on several servers at once, which are selected at runtime. Therefore, for each server that is selected, I spawn a new worker and assign its work. After everything is completed, the workers are deleted. I often get this error: BackgrounDRb::BdrbConnError in Multi com/indexController#uitvoeren Not able to connect Here''s the code that spawns the workers: (sorry for the linewraps) def uitvoeren # put selected servers in a session session[:selectedservers] = params[...

On Wed, May 31, 2023 at 11:27:12AM -0500, Mike Christie wrote: >On 5/31/23 10:15 AM, Mike Christie wrote: >>>> rcu would work for your case and for what Jason had requested. >>> Yeah, so you already have some patches? >>> >>> Do you want to send it to solve this problem? >>> >> Yeah, I'll break them out and send them later today when I

On Mon, Jun 05, 2023 at 01:57:30PM -0500, Mike Christie wrote: >If userspace does VHOST_VSOCK_SET_GUEST_CID before VHOST_SET_OWNER we >can race where: >1. thread0 calls vhost_transport_send_pkt -> vhost_work_queue >2. thread1 does VHOST_SET_OWNER which calls vhost_worker_create. >3. vhost_worker_create will set the dev->worker pointer before setting >the worker->vtsk

Hi ! First of all thanks for all your doing :) I also have this kind of segfault since the update : Dec 2 21:12:11 xxxxxxx dovecot: auth-worker: Error: *** Error in `dovecot/auth': double free or corruption (fasttop): 0x000055573bb99f70 *** Dec 2 21:12:11 xxxxxxx dovecot: auth-worker: Error: ======= Backtrace: ========= Dec 2 21:12:11 xxxxxxx dovecot: auth-worker: Error:

...e user of vhost_work_queue() is called in the ioctl context. See vhost_zerocopy_callback(). But since you want to limit the call before set_backend, another question comes, do we really need the dynamic attaching/creating in this case? How about a simple one ioctl that is used to build the queue to workers mapping instead? > Or we call flush from the file_operations release function > so the device is closed and can't race with ioctls. > > > > > Does this mean we need to hold vq mutex when doing the flush? (which > > seems not the case of vhost_scsi_tmf_resp_work()). &...

Signaling using normal kill(2) is preserved, but the master now prefers to signal workers using a pipe rather than kill(2). Non-graceful signals (:TERM/:KILL) are still sent using kill(2), as they ask for immediate shutdown. This change is necessary to avoid triggering the ubf (unblocking function) for rb_thread_call_without_gvl (and similar) functions extensions. Most notably, this f...

>>>>> "Luciano" == Luciano Mannucci <luciano at vespaperitivo.it> writes: > I'm trying to set up a dovecot server so that it authenticates local > user via /etc/passwd (I'm on a Freebsd 13.1) and via winbindd for > those that it cannot find localy. The samba suite is alive and well, > postfix gets happily mail from domain users and saves it with

Hello. We're setting up fts solr and want to have the solr server host be set per-user via UserDB. It looks like if a user connects and fts indexes mail, and then another user connects and indexes mail, indexer-worker is connecting to the first user's fts host: User1, hammer at rp-auth-test.com connects, does a SEARCH for the first time, indexer-worker gets UserDB settings and

...ker The latter patchset handles the review comment for the patches in thread to make sure that worker threads we create are accounted for in the parent process's NPROC limit. The patches are scheduled to be sent to Linus for 6.4. The patches in this patchset allow us to support multiple vhost workers per device. The design is a modified version of Stefan's original idea where userspace has the kernel create a worker and we pass back the pid. In this version instead of passing the pid between user/kernel space we use a worker_id which is just an integer managed by the vhost driver and we all...

I have a worker which wants to delete itself after it finishes its work- how can it do that? Best Regards, Danny Burkes

Since MiddleMan is an uninitialized constant in a worker how do you create new workers inside a worker? Thank You, Ben Johnson E: bjohnson at contuitive.com O: 800-341-6826 M: 817-229-4258 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/backgroundrb-devel/attachments/20060731/75a77fa6/attachment-0001.html