Pv5 - prl_disp_service consuming 99% cpu

  • Thread starter Thread starter Parallels User
  • Start date Start date
Hi,
first of all thanks everyone for your feedback - we do appreciate it.
What we need to deal with the problem is a bit more info from you.
When the situation is reproducing, and dispatcher is consuming CPU, open Terminal and paste the following commands:
1) prlsrvctl problem-report --send
2) sudo spindump -notarget 1 10 -file /tmp/samples.txt

- and give us both report ID and samples.txt file.

Thanks in advance,
Hanna
 
Activity monitor

I'm not sure you need to go messing around with sudo commands in Terminal to solve this.

I just had the same problem but cured it immediately by 'force quitting' the process in Activity monitor (ordinary Quit won't work).

1. Open Applications > Utilities > Activity Monitor.app

2. Ensure 'All Processes' is showing in the menu next to the Search/Filter bar

3. Type 'prl' in the filter bar (without the quote marks).

4. Choose the process in the window, and click 'Quit' in the taskbar.

5. Choose 'force quit' from the dialog box. You will need to enter an admin password.

(issue found with Parallels 6 on 10.6.8)
 
Last edited:
Actually that wasn't such a good idea, as it gave me this error msg when I tried to restart Parallels:

parallelserror.png


Back to Terminal then!
 
Hi Alto,

you still need Terminal to restore the Parallels environment (see reply from Parallels team previously posted on this thread):

"3. restore Parallels environment (if necessary):

sudo launchctl stop com.parallels.desktop.launchdaemon

wait a minute and perform:

sudo launchctl start com.parallels.desktop.launchdaemon"



Regards Phil
 
Alto said:
I'm not sure you need to go messing around with sudo commands in Terminal to solve this.
Click to expand...

It's not the way to solve something - just killing service from Activity Monitor. The issue will appear again and again. So we do need the information I've asked before to investigate the issue and find the way to solve it finally.
 
juzci said:
Hi,
first of all thanks everyone for your feedback - we do appreciate it.
What we need to deal with the problem is a bit more info from you.
When the situation is reproducing, and dispatcher is consuming CPU, open Terminal and paste the following commands:
1) prlsrvctl problem-report --send
2) sudo spindump -notarget 1 10 -file /tmp/samples.txt

- and give us both report ID and samples.txt file.

Thanks in advance,
Hanna
Click to expand...

Just an FYI - the first command you list just causes the prlsrvctl process to sit there and never complete. I had to amend it to "prlsrvctl problem-report --send --stand-alone" in order for it to finish.
I just did that, and the ID number is 12087622.

How do you want us to get the sample text file to you? It contains potentially private information, so I'm not willing to post it as an attachment to this forum.
 
sandro said:
ok, follow way is possible:

1. wait until pro_disp_service disappears from Activity Monitor after kill -ABRT sent
2. send problem report with command perform from Terminal:

prlsrvctl problem-report --send --stand-alone

3. restore Parallels environment (if necessary):

sudo launchctl stop com.parallels.desktop.launchdaemon

wait a minute and perform:

sudo launchctl start com.parallels.desktop.launchdaemon
Click to expand...

After months with no problem, this issue cropped up again. I followed your instructions, above. Problem report ID: 12224979

Isn't it time to fix this problem once and for all?!
 
Josh.Whitver, please look at your private messages.

cepheid, could you please attach the sample file too? In Terminal paste sudo spindump -notarget 1 10 -file /tmp/samples.txt and samples.txt file is what we need to investigate the issue.
 
Problem report

I've used the new instructions that juzci and Josh.Whitver have posted, and the problem report ID is 12292409.

@juzci, PM me so that I can get the samples.txt file to you.

Really hoping to get this resolved ASAP!
 
juzci said:
cepheid, could you please attach the sample file too? In Terminal paste sudo spindump -notarget 1 10 -file /tmp/samples.txt and samples.txt file is what we need to investigate the issue.
Click to expand...
Sorry, I forgot to create one of those. Looks like I'll have to wait until the next time this occurs. I'll keep you posted.
 
Thanks so much to everyone for samples and reports!
All the data we have now is about Parallels Desktop ver.6 - is there anyone with same problems on PD7 for us to be sure what we are looking for?
 
It happens with Windows running too

I'm directing this message to Juzci or anyone else from the Parallels Team following this thread:

Today, for the first time, I had prl_disp_service do its 99% thing with Windows running. Naturally, I panicked when Temperature Monitor Lite reported a CPU temperature of 100 Celsius and forgot to run the error snapshot you have requested. I will try to remember next time. I should note that Parallels seemed to hang after the Windows desktop disappeared and I had to manually quit it.

I noticed on your most recent post that you asked if anyone was having this problem with Parallels 7. I'm running Pv6, but when I read through this entire message thread a few months ago, I noticed people with Pv7 complaining about this problem, so I didn't upgrade. All of us seem to have one thing in common: we have Snow Leopard 10.6.8.

I'm both unable (I lack a Time Machine backup drive) and reluctant to downgrade to 10.6.7 (because I had graphics card glitches with 10.6.3-7). I also do not want to upgrade to Lion or Mountain Lion because my Mac's a bit older (2009, shipped with Leopard).

Please do what you can to fix this problem. The Terminal commands that reset prl_disp_service have allowed me to stick with Parallels, but my patience is getting weaker, and I hear VirtualBox is getting better...
 
This thread has been running for over 2 years now and still there is no solution for this problem other than having to restart the Parallels service within the terminal, and I don't consider that a fix. This is not even a version-dependent problem that has only recently cropped up. I definitely won't be "upgrading".
 
I have downloaded last update however when parallels is working my mac slowdown. I cant do anything because of memory problem (only 10-50mb memory left out of 2GB)
I have MBP 13" early 2011 (4GB RAM) with Mac OS 10.7.3 and win 7 (parallels 7).
Unfortunately I could not use parallels because if above slowdown problem.
What should I do?
below you can find screenshot showing activity monitor.
http://d.pr/h6at

Best regards,
 
I've been suffering this bug for about a year now. It's a bit annoying.

I am willing to upgrade to Parallels 7 now, if it fixes the issue. However, I don't want to upgrade if it won't. If I upgrade to PD7, can you confirm that this problem will be fixed?
 
I can assure you that upgrading to PD7 does NOT help to alleviate this problem, because when I upgraded to PD7, this very problem STARTED for me. Now all of my virtual machines run so slow they are essentially unusable.
 
OK so updating to v7 is not guaranteed to fix this issue... so I don't think I will do that.

Parallels team - could you give us an update on what's causing this issue and timescale/plan for a fix?

I could give Fusion a try I guess but really would rather stay with Parallels. So I think I will hang on for a month or so to see what Parallels come up with...
 
Garote,
could you please send us some data about your issue on PD7?
When the situation is reproducing, and dispatcher is consuming CPU, open Terminal and paste the following commands:
1) prlsrvctl problem-report --send
2) sudo spindump -notarget 1 10 -file /tmp/samples.txt

- and give us both report ID and samples.txt file.

Thanks in advance,
Hanna
 
You must log in or register to reply here.
Back
Top