I have been able to get CW for ColdFire ver. 7.2 EVALUATION version to run on several PCs, but not on my main workstation PC. The Installation seems to work fine (no errors are reported). But when I try to run CodeWarrior, it starts to bring up the background of the IDE, then immediately "crashes". Windows opens a dialog box that reports that:
Integrated Development Environment has encountered a
problem and needs to close. We are sorry for the inconvenience. (etc, etc.)
And, this dialog box asks if I want to send a problem report to Microsoft or not.
Now, this is the 30-day EVAL copy (I haven't loaded any "dongle drivers" or license files yet.) The IDE just
won't run...it just blows away within 3 or 4 seconds after starting.
Every PC I was able to get CW to run on was a uniprocessor PC. My "workstation" PC is a Intel QUAD processor,
running XP Pro (32 bit). Does CW not run on multi-core PCs?
One other possibility: my "workstation" PC has many programs on it that are managed by FlexLM. Many of these programs are 10 years old or older, so there are many copies of FlexLM and lmgrd.xxx in that PC. Is it possible that lmgrd might be fighting with older versions? If so, how can I debug / fix this? FlexLM is a DISASTER when trying to run many licensed programs on the same PC. Can anyone help me with this>?
已解决! 转到解答。
I just wasted a couple of days tracking down something that sounds like the same problem:
Running CW for Coldfire 7.2 Special Edition on Windows XP pro. It would crash every few minutes, giving me this in the event log:
Event Type: Error Event Source: Application Error Event Category: None Event ID: 1000 Date: 5/8/2010 Time: 1:19:54 AM User: N/A Computer: JCG-LAPTOP Description: Faulting application ide.exe, version 5.9.0.5293, faulting module lmgr11.dll, version 0.0.0.0, fault address 0x0007f841. For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp. Data: 0000: 41 70 70 6c 69 63 61 74 Applicat 0008: 69 6f 6e 20 46 61 69 6c ion Fail 0010: 75 72 65 20 20 69 64 65 ure ide 0018: 2e 65 78 65 20 35 2e 39 .exe 5.9 0020: 2e 30 2e 35 32 39 33 20 .0.5293 0028: 69 6e 20 6c 6d 67 72 31 in lmgr1 0030: 31 2e 64 6c 6c 20 30 2e 1.dll 0. 0038: 30 2e 30 2e 30 20 61 74 0.0.0 at 0040: 20 6f 66 66 73 65 74 20 offset 0048: 30 30 30 37 66 38 34 31 0007f841 0050: 0d 0a ..
As you see, lmgr11.dll is fingered as the culprit. It apparently woke up after perhaps minutes and blew up. I determined that this only happened on a particular computer, and only if the project file resided on a particular disk partition. Then I monitored file activity with filemon.exe, and learned that after 5 minutes or so, lmrg11.dll wakes up and scans the root directory of the partition that the project file is on. It opens every file and directory in the root partition.
On the system in question, there is one file in the root directory that has a corrupted last-modification timestamp. Windows Explorer shows a timestamp of Wednesday, November 24, 2094, 8:13:02 PM. The os.stat() function in python returns a large negative number st_mtime=-353477714, which the various datetime functions refuse to parse. I have no idea why this file is corrupted that way, but it has apparently been there for years.
When lmgr11.dll attempted to touch that file with the bad timestamp, it crashed. Someone needs to feed this back to flexlm tech support.
My solution was easy enough -- I moved that file out of the root directory and will eventually update the timestamp once I'm done looking at this problem. IAnyone experienceing a problem like this should look for a corrupted timestamp on a file somewhere on the disk partition.
This issue raises a couple of questions in my mind:
Yes thanks for the post. I ended up re-installing Windows from CD and that fixed the problem. Still an issue in CW 10.6. If you have a crash in file LMGR11.dll this is the problem to look for.
I have very similar problem on one of my workstations. It's uniprocessor, so I don't think number of CPUs matters. Error message points to lmgr11.dll file, which is some version of Flex license manager. I do not have any lmgr*.dll files other then included with CW 7.2
It appears that lmgr11.dll is what is blowing away, although I don't
know why. While I have other programs in this PC that are protected by
FlexLM, none of them are running at the same time. (I've checked; no other
version of lmgrxx.xxx is running in the background). So, why is lmgr11.dll
blowing up? There doesn't seem to be any reason for it
Is there any way to test lmgr11.dll by itself?.
I just wasted a couple of days tracking down something that sounds like the same problem:
Running CW for Coldfire 7.2 Special Edition on Windows XP pro. It would crash every few minutes, giving me this in the event log:
Event Type: Error Event Source: Application Error Event Category: None Event ID: 1000 Date: 5/8/2010 Time: 1:19:54 AM User: N/A Computer: JCG-LAPTOP Description: Faulting application ide.exe, version 5.9.0.5293, faulting module lmgr11.dll, version 0.0.0.0, fault address 0x0007f841. For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp. Data: 0000: 41 70 70 6c 69 63 61 74 Applicat 0008: 69 6f 6e 20 46 61 69 6c ion Fail 0010: 75 72 65 20 20 69 64 65 ure ide 0018: 2e 65 78 65 20 35 2e 39 .exe 5.9 0020: 2e 30 2e 35 32 39 33 20 .0.5293 0028: 69 6e 20 6c 6d 67 72 31 in lmgr1 0030: 31 2e 64 6c 6c 20 30 2e 1.dll 0. 0038: 30 2e 30 2e 30 20 61 74 0.0.0 at 0040: 20 6f 66 66 73 65 74 20 offset 0048: 30 30 30 37 66 38 34 31 0007f841 0050: 0d 0a ..
As you see, lmgr11.dll is fingered as the culprit. It apparently woke up after perhaps minutes and blew up. I determined that this only happened on a particular computer, and only if the project file resided on a particular disk partition. Then I monitored file activity with filemon.exe, and learned that after 5 minutes or so, lmrg11.dll wakes up and scans the root directory of the partition that the project file is on. It opens every file and directory in the root partition.
On the system in question, there is one file in the root directory that has a corrupted last-modification timestamp. Windows Explorer shows a timestamp of Wednesday, November 24, 2094, 8:13:02 PM. The os.stat() function in python returns a large negative number st_mtime=-353477714, which the various datetime functions refuse to parse. I have no idea why this file is corrupted that way, but it has apparently been there for years.
When lmgr11.dll attempted to touch that file with the bad timestamp, it crashed. Someone needs to feed this back to flexlm tech support.
My solution was easy enough -- I moved that file out of the root directory and will eventually update the timestamp once I'm done looking at this problem. IAnyone experienceing a problem like this should look for a corrupted timestamp on a file somewhere on the disk partition.
This issue raises a couple of questions in my mind:
thanks for the post, it's still very relevant. I ran into this yesterday with CW 10.5 on win XP SP2. My addition is that it could also be a file name issue. I had my workspace on a drive shared with OSX (running CW in parallels), and the linker (which launches lmgr11.dll) was searching the root of the workspace drive, and found a name with parentheses in it and crashed. Lmgr11.dll itself wasn't scanning the directories it was the parent. So my advice, also look out for filenames (probably only relevant if you're on a non DOS/FAT32/NTFS drive).
AH! You're right! lmgr11.dll does apparently scan several directories for file dates
(to see if your 30-day evaluation period is over). One place it goes is to c:\Windows .... and
it scans every file. In my case, I had several files with the name (whatever).ocx ..... with the
file Type listed as "ActiveX Control". But, the file dates were not visible. (See the attached
screenshot). When I checked the "properties" of these files, they had file dates in the year 1617.
Nice.
I downloaded a utility that allowed me to change file dates. (And, you have to get them ALL....
including the file created date..... ALL dates have to be valid, and recent. I'm assuming > 1980
or so.) Once fixed up, CodeWarrior works just fine.
Thanks also for the tip about filemon.exe. I found that this functionality is included in
the "Process Monitor" from sysinternals. http://www.sysinternals.com
To find files that cause lmgr11.dll to choke:
1) Download the process monitor, start it
2) Start CodeWarrior.
3) When lmgr11.dll "crashes" .... look at the BOTTOM of the screen, to see
the last files that lmgr11.dll was trying to open. One of them will have a corrupt
file date.
I have reported this error to the CodeWarrior developers..... they will report it back
to Flexera, and try to get this "bug" cleaned up.
--Bill
Great troubleshooting. I had 3 OCXs, all with numbered names and invalid dates. I did a registry search for all of them and they were not in the registry, so I am pretty sure they are not valid or used. They are much easier to archive than to change both the created and modified dates.
Niice thanks I just lost three working days to that one - the real joy is that you can uninstall+reinstall the whole f'ing eclipse package (wasting plenty of time) and it still screws you.
For the record my problem was my machine started up one day thinking it was the year 2080 (dodgy clock battery), and I got the file
c:\Windows\PFRO.log
created by some regular windows (7) process, with a date in the year 2080.
I just deleted that, and yay, works again. Of course now I lost three work days to another stupid compiler issue (also see "Linker silently corrupts INCBIN data" etc).
Grr. You'd think there'd be a big sign somewhere to warn people of this issue wouldn't you? Hey maybe the license manager DLL gets less f'witted or perhaps just gets removed...
Congratulations, it seems you've found the source of the problem. I've check my workstation for files with strange datestamps and I've found nothing. Anyway, there was one file in C:/windows dated 1980-01-01 so to be sure I changed this to 1981-01-01. And now CW7.2 starts flawlessly !
There is nothing malicious for the license manager to look for timestamps of various files on the disk, it tries to detect cases when calendar on the PC was set backwards. But I would prefer the software is tested before releasing it to the customers