config fails: Parallels 2.2 on Fedora 7 with 2.6.22 kernel

Please indicate which release of Fedora you use with Parallels 2.2 and rate stability

  • Fedora 7: Rock Solid

    Votes: 0 0.0%

  • Fedora 7: Somewhat Unstable

    Votes: 2 66.7%

  • Fedora 6: Rock Solid

    Votes: 0 0.0%

  • Fedora 6: Somewhat Unstable

    Votes: 0 0.0%

  • Fedora 5: Rock Solid

    Votes: 0 0.0%

  • Fedora 5: Somewhat Unstable

    Votes: 0 0.0%

  • Fedora 5/6/7: Highly Unstable / Not usable

    Votes: 0 0.0%

  • Other: Rock Solid (ie: Redhat Enterprise)

    Votes: 0 0.0%

  • Other: Somewhat Unstable

    Votes: 0 0.0%

  • Other: Highly Unstable / Not usable

    Votes: 1 33.3%
  • Total voters
    3
The patched sources still are not compiled under debian-i386/unstable with kernel 2.6.22.1 and gcc 4.2.1 (I also tried gcc 4.1.3 with no luck).
The following errors break the compilation:

/usr/lib/parallels/drivers/drv_main/mm/pages.c: In function 'mmLockPages':
/usr/lib/parallels/drivers/drv_main/mm/pages.c:51: error: dereferencing pointer to incomplete type
/usr/lib/parallels/drivers/drv_main/mm/pages.c:63: error: dereferencing pointer to incomplete type
/usr/lib/parallels/drivers/drv_main/mm/pages.c:64: error: dereferencing pointer to incomplete type
/usr/lib/parallels/drivers/drv_main/mm/pages.c:66: error: dereferencing pointer to incomplete type
/usr/lib/parallels/drivers/drv_main/mm/pages.c: In function 'mmKGetPageByAddr':
/usr/lib/parallels/drivers/drv_main/mm/pages.c:176: error: 'init_mm' undeclared (first use in this function)
/usr/lib/parallels/drivers/drv_main/mm/pages.c:176: error: (Each undeclared identifier is reported only once
/usr/lib/parallels/drivers/drv_main/mm/pages.c:176: error: for each function it appears in.)
/usr/lib/parallels/drivers/drv_main/mm/pages.c: In function 'mmUGetPageByAddr':
/usr/lib/parallels/drivers/drv_main/mm/pages.c:198: error: dereferencing pointer to incomplete type
/usr/lib/parallels/drivers/drv_main/mm/pages.c:200: error: dereferencing pointer to incomplete type
/usr/lib/parallels/drivers/drv_main/mm/pages.c:207: error: dereferencing pointer to incomplete type
/usr/lib/parallels/drivers/drv_main/mm/pages.c:213: error: dereferencing pointer to incomplete type
/usr/lib/parallels/drivers/drv_main/mm/pages.c: In function 'mmGetPages':
/usr/lib/parallels/drivers/drv_main/mm/pages.c:235: error: dereferencing pointer to incomplete type
 
1nept said:
The patched sources still are not compiled under debian-i386/unstable with kernel 2.6.22.1 and gcc 4.2.1 (I also tried gcc 4.1.3 with no luck).
The following errors break the compilation:

/usr/lib/parallels/drivers/drv_main/mm/pages.c: In function 'mmLockPages':
/usr/lib/parallels/drivers/drv_main/mm/pages.c:51: error: dereferencing pointer to incomplete type
Click to expand...

I see. I's all because of missing header 'linux/sched.h' in pages.c.
I've posted 2 patches. Have you tried last one?
 
hi aak,

i've been having lots of problems lately - my system is very unstable whenever i try to run parallels. lots of weird things happen - like right now the shift keys on the keyboard are not recognized in linux after parallels crashed. my system crashes now at least once a day unless i try not to use parallels, but today there have been many crashes.

just now parallels died; i looked in /var/log/messages and found this;

Aug 24 00:50:56 dev3 kernel: [ 9594] Application process unexpectly died (/usr/lib/parallels/drivers/drv_main/device.c:106)
Aug 24 00:50:56 dev3 kernel: device eth0 left promiscuous mode

in case it helps;

[azer@dev3 ~]$ uname -s -r -v -m -p -i -o
Linux 2.6.22.1-41.fc7 #1 SMP Fri Jul 27 18:10:34 EDT 2007 i686 i686 i386 GNU/Linux

the cpu is core2 duo.

by the way; many times my entire system dies. one day i tried to log in from another computer via ssh and was able even though the workstation was frozen completely in every other sense. i then tried to list running tasks with ps auxwf and the session froze. over time i came to realize that there were serious memory corruption issues that appear even if parallels does not crash.

one of the microsoft bsod screens mentioned interrupt issues. another microsoft message - after rebooting the system and restarting parallels - talked about memory corruption but most of them talk about a problem with the drivers. trying to log-in to the microsoft virtual pc using remote desktop seems to work fine but using nx to log-in to the x-windows session seems to cause parallels to be even more unstable.

i realize that all this might easily come out of the driver changes that we've made but i have no real hope of getting a handle on any part of it. if you have the necessary debugging tools available, as you can imagine, i'd appreciate very much any updates you can provide.

thanks again,
have a great day,
take care,
sam.
 
PS: I was using the Flicker upload tool the day that the system was crashing constantly. The problems were probably related to heavy network traffic as the images being uploaded were on a share. HTH, --Sam.
 
Hi Aak,

I've been using Parallels less lately - just booting my WinXP VM when I need it instead of running it all the time. Both my Linux workstation and the Parallels VM crash much less lately - no crashes for a few days now (instead of about one a day before.) The keyboard also disconnects much less now.

My Parallels VM just crashed now. Enterprise Architect and MS Outlook were open but I was busy writing code in a Linux shell - nothing really going on. A BSOD popped up and a memory dump began (I changed my WinXP error handling settings as per a previous instruction from Microsoft.) Sorry I wasn't fast enough to capture the screen - the message indicated a problem with interrupts. When I rebooted, I got a Windows Error Reporting message saying again that the problem was due to a device driver.

So, all in all, I'm confident that the recent changes to the kernel are causing some instability with the Parallels drivers. As I really do need my WinXP VM I'm thinking of going back to an older kernel.

Mostly, though, I'd appreciate any ETA you might have for the next update for Parallels. It's been a while since your last update and I understand you have lots of interesting new features in your current code base. Please concentrate on reliability - the feature set is nice but it's not so important relative to basic functionality.

Please offer some indication as to the delay remaining before your next release - I'll use the information as a guide only; I won't hold you to it.

Thanks again,
All the best,
Take Care,
Sam.
 
Im not running 64-bit and recieve the same error, can somebody help please?

Info:
Im running FC7 with the latest updates with a fluxbox wm incase it matters
intel p4

my problem lies with both the manager.c and prlnet.c
and the manager.c problem was fixed by putting the new header line as indicated
however i cannot get the patch to work in prlnet.c
im not new to command line or linux but have never needed to use a patch, so i tried exactly as stated before, to no avail.

the patch hangs after inputting the command in the right place and right filename.
i ran the patch with sudo, and su for hopes of a difference but here is the run through:

5yhynac.jpg


am i doing something wrong?

(ps i pressed ctrl+c to terminate, it didn't end on its own, it hung after patch
 
Last edited:
48hrs straight with no crash...

Hi Folks,

Thanks for the recent update! I just installed it today.

I wanted you (Parallels staff) to know that I installed a few days ago a Kernel update (2.6.22.5-76.fc7) along with a large number of other updates. After that I tried running my WinXP VirtualPC for a while and it ran fine for almost 48 hours. I had begun to suspect that perhaps some of the problems were being caused by media player software running in Linux. Perhaps there was also something going wrong in one of the previous kernel updates.

Thanks again,
Take Care,
Sam.
 
Hi Folks,

Things are much better in a sense these past couple of days. The Clipboard works better - I can copy something now from the Virtual PC and paste it in Linux. Until today the Virtual PC did not crash except at night when I wasn't using it (probably due to scheduled tasks like Windows Defender scans.)

Just now, though, I was listening to CNN Live and got a BSOD. Then, after restarting and taking a snapshot of the Microsoft Error Response page another BSOD appeared. The Microsoft Error Response page is the same one that was shown for a number of crashes in the past. Screen captures are here:

http://www.flickr.com/gp/11927456@N02/14iB0T

Again, it seems to me that the last batch of updates (both the updated Linux Kernel and the updated Parallels workstation) corrected a number of problems. Keep it up! We want a rock-stable environment in the end and it seems we're closer now.

For now, please let me know if you have any links to pages that contain information on how to collect more debugging information when something crashes.

Thanks,
Take Care,
Sam.
 
Last edited:
Fedora Updates Available

Hello all,

The fedora people have posted a number of updates to F7.

It dawned on me that xen, qemu and kvm modules might conflict with Parallels so I removed them on my system. No harm done but it's too soon to tell if there's any major benefit.

Best wishes to all,
Take Care,
Sam.
 
There were too many crashes per day with the 2.6.22.9-91.fc7 kernel; I'm going back to 2.6.22.5-76.fc7. In case it helps: The error messages from Microsoft are always about the device drivers lately. --Sam.
 
In case it's of any use to <b>Parallels developers</b>, please see the last photo in the set below. The error message for most of the BSOD screens is PFN List Corrupt. This last one is IRQL Not Less or Equal. So far all the error response web pages from Microsoft say that the problem is with a device driver.

http://www.flickr.com/gp/11927456@N02/14iB0T

Thanks,
Take Care,
Sam.
 
Tried running as root:

[root@dev3 ~]# parallels &
[1] 25674
[root@dev3 ~]# Xlib: extension "RANDR" missing on display ":0.0".
deleting folders!!
deleting folders!!
CApp::VmCallback(): Unknown message code: 0x16
CApp::VmCallback(): Unknown message code: 0x16
CApp::VmCallback(): Unknown message code: 0x16
deleting folders!!
deleting folders!!
CApp::VmCallback(): Unknown message code: 0x16

[1]+ Done parallels
[root@dev3 ~]#


No software was running in the WinXP client although I did try to connect the USB interface for my pocket PC (it didn't actually sync but the USB device was supposedly connected in Windows.) While I was busy with something else for a few minutes, something went wrong. A BSOD appeared but was not able to write the memory dump. This never happened before.

Note that I had removed the Parallels tools (and re-installed the Parallels Display Driver) as per some notes on the Microsoft web site.

HTH,
Sam.
 
I've added a Poll because I'm curious to know if there's a release of Fedora that Parallels can run on in a highly stable manner. In general, this is not a good place to ask such questions because the people who will answer are those searching for help with a problem. What we really want is a general user survey or responses from Beta testers only.
 
Hi, I recently purchased and installed Parallels Workstation. I install a 2112 release initially and applied the patch. The application worked fine till I updated my Nvidia driver and applied the auto update from Parallels.

Here's my log:

<snip>
make[2]: Leaving directory `/usr/lib/parallels/drivers/drv_main'
cd hypervisor/ && make clean && cd ..
make[2]: Entering directory `/usr/lib/parallels/drivers/hypervisor'
make[2]: Leaving directory `/usr/lib/parallels/drivers/hypervisor'
cd drv_net/linux/ && make clean && cd ..
make[2]: Entering directory `/usr/lib/parallels/drivers/drv_net/linux'
rm -rf *.o *.ko .*.cmd *.mod.c .tmp_versions
make[2]: Leaving directory `/usr/lib/parallels/drivers/drv_net/linux'
cd drv_virtualnic/ && make clean && cd ..
make[2]: Entering directory `/usr/lib/parallels/drivers/drv_virtualnic'
rm -rf *.o *.ko .*.cmd *.mod.c .tmp_versions
make[2]: Leaving directory `/usr/lib/parallels/drivers/drv_virtualnic'
cd drv_main/ && make KSRC=/lib/modules/2.6.23.1-10.fc7/build && cd ..
make[2]: Entering directory `/usr/lib/parallels/drivers/drv_main'
make -C /lib/modules/2.6.23.1-10.fc7/build SUBDIRS=/usr/lib/parallels/drivers/drv_main SRCROOT=/usr/lib/parallels/drivers/drv_main modules
make[3]: Entering directory `/usr/src/kernels/2.6.23.1-10.fc7-i686'
CC [M] /usr/lib/parallels/drivers/drv_main/vmmain.o
In file included from /usr/lib/parallels/drivers/drv_main/vmmain.c:29:
/usr/lib/parallels/drivers/drv_main/vmmain.h:196: error: expected specifier-qualifier-list before ‘kmem_cache_t’
/usr/lib/parallels/drivers/drv_main/vmmain.c: In function ‘init_module’:
/usr/lib/parallels/drivers/drv_main/vmmain.c:82: error: ‘struct <anonymous>’ has no member named ‘totalMem’
make[4]: *** [/usr/lib/parallels/drivers/drv_main/vmmain.o] Error 1
make[3]: *** [_module_/usr/lib/parallels/drivers/drv_main] Error 2
make[3]: Leaving directory `/usr/src/kernels/2.6.23.1-10.fc7-i686'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/usr/lib/parallels/drivers/drv_main'
make[1]: *** [vmmain] Error 2
make[1]: Leaving directory `/usr/lib/parallels/drivers'
make: *** [build] Error 2

Any ideas what went wrong?

I on F7 with kernal 2.6.23 (as you can see above)
 
Is anybody else having trouble running parallels-config under kernel 2.6.23 r3?

After patching:

patch -p0 < ~/parallels-2.2.2112-linux-2.6.22.patch
patching file drivers/drv_main/mm/manager.c
patching file drivers/drv_net/linux/prlnet.c

./parallels-config results in an error log:

:
/usr/lib/parallels/drivers/drv_main/vmmain.c: In function 'init_module':
/usr/lib/parallels/drivers/drv_main/vmmain.c:82: error: 'struct <anonymous>' has
no member named 'totalMem'
make[4]: *** [/usr/lib/parallels/drivers/drv_main/vmmain.o] Error 1
make[3]: *** [_module_/usr/lib/parallels/drivers/drv_main] Error 2
make[3]: Leaving directory `/usr/src/linux-2.6.23-gentoo-r3'
 
You must log in or register to reply here.
Back
Top