Hi Wayne
We've seen this problem here. I thought it had been discussed in the mailing list before but after just having a quick search I can't find anything.
I'm convinced that it has something to do with peak picking/assigning (ie the more peaks you do the sooner it crashes). As mentioned before analysis does slow up before the crash and if you save and restart it seems okay again.
We're using Suse 11
Analysis 2.0.7 - pre-compiled 64bit
Tk mode
Python 2.5
Tk8.5
X.Org X Server 1.4.0.90
Here are a few of the error messages I've had in the past when I've been too lazy to restart.
X Error of failed request: BadIDChoice (invalid resource ID chosen for this connection)
Major opcode of failed request: 154 (RENDER)
Minor opcode of failed request: 4 (RenderCreatePicture)
Resource id in failed request: 0x1c926e5
Serial number of failed request: 146178515
Current serial number in output stream: 146180230
X Error of failed request: BadIDChoice (invalid resource ID chosen for this connection)
Major opcode of failed request: 53 (X_CreatePixmap)
Resource id in failed request: 0x2eae6a5
Serial number of failed request: 116861125
Current serial number in output stream: 116861141
X Error of failed request: BadIDChoice (invalid resource ID chosen for this connection)
Major opcode of failed request: 154 (RENDER)
Minor opcode of failed request: 4 (RenderCreatePicture)
Resource id in failed request: 0x2e9cec6
Serial number of failed request: 143505743
Current serial number in output stream: 143505749
X Error of failed request: BadIDChoice (invalid resource ID chosen for this connection)
Major opcode of failed request: 1 (X_CreateWindow)
Resource id in failed request: 0x2c43772
Serial number of failed request: 190959780
Current serial number in output stream: 190959804
X Error of failed request: BadIDChoice (invalid resource ID chosen for this connection)
Major opcode of failed request: 154 (RENDER)
Minor opcode of failed request: 4 (RenderCreatePicture)
Resource id in failed request: 0x38d864f
Serial number of failed request: 137823668
Current serial number in output stream: 137823684
X Error of failed request: BadIDChoice (invalid resource ID chosen for this connection)
Major opcode of failed request: 1 (X_CreateWindow)
Resource id in failed request: 0x2c5ef48
Serial number of failed request: 271529015
Current serial number in output stream: 271539482
Hope it helps
Michelle
Dr Michelle Rowe
Assistant Manager of the Biomolecular NMR Facility
School of Biosciences
University of Kent
Canterbury
Kent
CT2 7NJ
United Kingdom
Tel: (+44) 1227 827117
Email: [log in to unmask]
-----Original Message-----
From: CcpNmr software mailing list [mailto:[log in to unmask]] On Behalf Of Wayne Boucher
Sent: 07 October 2009 18:53
To: [log in to unmask]
Subject: Re: ccpn crash
So, it seems that Tcl/Tk 8.4 did not solve this crashing problem on Red
Hat. So my guess now must be that this is not related to aliased fonts
(etc.) but something else. Has anyone (so other than Stuart on Red Hat
and Philipp on OSX10.4) else had this crashing problem (with the
BadIDChoice / RenderCreatePicture message)? And has anyone seen it in
OpenGL mode?
Wayne
On Mon, 5 Oct 2009, Wayne Boucher wrote:
> Hello,
>
> That's the same as my Mac, which is interesting. On the other hand, I don't
> use the precompiled Analysis day-to-day but instead one which uses (for
> historic reasons) Tk 8.4, so that is one difference. I would recommend first
> trying out OpenGL to see if that helps. But it might not, since everything
> aside from the spectra canvases uses Tk. Perhaps I should also do a Tk8.4
> pre-compiled release for the Macs, like I did for Linux. (So it would be
> good to know if this really is a Tk8.4/Tk8.5 issue.)
>
> Wayne
>
> On Mon, 5 Oct 2009, Philipp Marchand wrote:
>
>> Hi,
>>
>> OS version 10.4.11
>> Analysis 2.1 precompiled
>> the X version is X11 1.1.3 - Xfree86 4.4.0
>> Tk mode
>>
>> Philipp
>>
>> 2009/10/5 Wayne Boucher <[log in to unmask]>
>> Hello,
>>
>> The same kind of error has been observed on Linux. I had come across
>> some webpage which claimed this has something to do with aliased
>> fonts
>> in Tk8.5, so see:
>>
>> https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind0904&L=CCPNMR&P=R8902&X=200DBA4F415$
>>
>> But I just found another post which claims it might not be to do with
>> aliased fonts (so in that post above I might have conflated two
>> issues) but instead just with something called xcb:
>>
>> http://lists.macosforge.org/pipermail/xquartz-dev/2009-April/002252.html
>>
>> So I guess it's worth asking various questions to see if we can
>> narrow
>> it down.
>>
>> Which OSX are you using (so 10.4/10.5/10.6)?
>>
>> Which version of Analysis are you using? (It looks like the
>> pre-compiled Intel OSX one for 2.1. So that does use Tk8.5.)
>>
>> What is your X version? (So look in "About X11" under the X app.)
>>
>> Is this in Tk mode or in OpenGL mode or both?
>>
>> Wayne
>>
>> On Mon, 5 Oct 2009, Philipp Marchand wrote:
>>
>> Hello,
>>
>> it happened several times now that CCPN closes all of a
>> sudden. Fortunately, the crash is
>> announced in a way that the program gets slower and slower
>> before it closes.
>>
>> The following error message appears in the xterm console:
>>
>> X Error of failed request: BadIDChoice (invalid resource
>> ID chosen for this connection)
>> Major opcode of failed request: 151 (RENDER)
>> Minor opcode of failed request: 4 (RenderCreatePicture)
>> Resource id in failed request: 0x82b6cd
>> Serial number of failed request: 16950659
>> Current serial number in output stream: 16950734
>>
>> I'm using a MacBookPro with everything a Mac fan can
>> imagine ;)
>> Any suggestions?
>>
>> Thanks,
>> Philipp Marchand
>>
>>
>>
>>
>>
>
|