| Quick navigation: | Home | Site Map || References | Biography || Copyright | Other copyright | Contact us | Advert | | |
Re: [ccp4bb] Refmac Problem - Permission Denied |
||
- Protein crystallographyMain steps:- Protein purification- Crystallisation Special:- Programs for crystallography- X-ray detectors Basic tutorials:- Chemistry- Protein - Peptide - Amino Acids Xtal community:- CCP4BB |
CCP4bb navigationCCP4bb <-- 1999 <-- November 1999 <-- 30 November 1999Subject: Re: Refmac Problem - Permission Denied From: David Jones David {- dot -} Jones {- at -} UCDENVER {- dot -} EDU Date: 2009-09-02 The problem is that every directory in lib/data/monomers has to be world "writeable" in order to use nfsv4 for refmac to run. Something that clearly I don't want to do... And why just the monomers directory.... is refmac trying to write something, or is it a broken nfslock I can easily fix this with an NFS v3 mount. Which is what I have done. Just wondered if anyone else has experienced this. On Sep 2, 2009, at 4:45 AM, Adam Ralph wrote: Hi David, Do all the client machines have NFSv4? You can force a mount to use NFSv3 by using "nfsvers=3" as a mount option. I guess that the server might have something similar. That would allow you to make more secure directory settings. I am not sure how to fix the problem, it could be a bug. Certainly permissions on the lower directories are important, 'x' allows you to `cd` into that directory. Adam On Tue, 1 Sep 2009, David Jones wrote: > Hi Adam > > I can write to the server, I can edit the files. > overrode default permission, mounted as rw etc... > > Turns out every directory under monomers has to be owned by the user > or have rwx permission for everyone set. > I have made sure extended access control lists are turned off, > mounted with insecure locking etc etc. > nfs3 works without any problem with default permissions... > not sure what I am missing here.. > > > David > > ------------------------------------------------ > David N. M. Jones, Ph.D. > Associate Professor > Dept. of Pharmacology > Univ. of Colorado Denver > 12801 East 17th Ave > MS 8303, PO Box 6511 > Aurora, CO 80045 > > e-mail david.jones@ucdenver.edu > Tel. (303) 724-3600 > Cell (303) 916-7246 > > > > > > > On Sep 1, 2009, at 6:47 AM, Adam Ralph wrote: > > > Hi David, > > You have probably sorted this by now but can you save any > files remotely to the server? The nfs disk will have mount > options that could make it readonly, these will supercede > any local permissions. Check /etc/fstab. > > Adam > > > > On Sat, 29 Aug 2009, David Jones wrote: > >> I would appreciate if any one has any ideas on this >> >> >> I have a strange NFS related problem, if anyone has any ideas... >> >> I am running refmac with the ccp4 directories mounted on an nfs >> share (v4) mounted under /usr/local >> >> I get the following error... >> >> #CCP4I TERMINATION STATUS 0 Last system error message: Permission >> denied Refmac_5.5.0072: Open failed: File: /usr/local/ccp4-6.1.1/ >> lib/data/monomers/mon_lib_ind.cif >> >> >> However, to test this I have changed the permissions to 777 and uid >> and gid to the user's account. >> The user can see the file, and can even edit the file on the remote >> server. >> and yet it still fails. If I run refmac from the users account on >> the nfs server directly, then it runs with no problem. >> >> This is running under CentOS 5.3, fully patched etc, with SE linux >> turned off >> >> >> >> ------------------------------------------------ >> David N. M. Jones, Ph.D. >> Associate Professor >> Dept. of Pharmacology >> Univ. of Colorado Denver >> 12801 East 17th Ave >> MS 8303, PO Box 6511 >> Aurora, CO 80045 >> >> e-mail david.jones@ucdenver.edu >> Tel. (303) 724-3600 >> Cell (303) 916-7246 >> > CCP4bb navigationCCP4bb <-- 1999 <-- November 1999 <-- 30 November 1999 |
|
| ProteinCrystallography.org: Copyright 2006-2010 by Quid United Ltd |