Grigory, Based on your tone, i won't even reply in details. It seems that you know better OSCAR than me, i can assure you that this is normal is /etc/fstab is not in the image and if the initrd is not in the image either; SystemConfigurator is supposed to deal with that at the end of the node deployment. SystemConfigurator may have an issue but SystemConfigurator is not specific to OSCAR (i may implement a work around if someone manges to send me all the needed details to actually do so efficiently.
And BTW __oscar-6.x is not yet ready for production__. Since you give the attitude, I won't reply to any of your emails from now. Regards, ----- Mail Original ----- De: "Grigory Shamov" <ga...@yahoo.com> À: oscar-users@lists.sourceforge.net Envoyé: Mardi 28 Avril 2009 13h36:42 GMT -05:00 USA/Canada - États de l'Est Objet: Re: [Oscar-users] 6.0.2, 6.0.3 and broken initrd Hi Geoffroy, I just want to say that I had the same issue -- with both unstable and stable OSCAR 6.0's, with CentOS 5.2 and 5.3. I've seen messages about musing /etc/fstab during the corresponding OSCAR stages, and kernels fail to find root, no matter whether YUOK kernel checkbox is on in the OSCAR wizard. I dont think that the attitude "its not an OSCAR problem" is very productive. There is a problem, and if to solve it a user has to manually work with SystemImager -- what does he need the OSCAR for? My conclusion was that the OSCAR 6 is not yet stable to use for now, and there is no manpower for its developers to fix it, so I switched to another platform for now. I couldnt afford any more downtime for my hardware. May be for my next cluster I will try OSCAR again, for I like it (have used with success in the past 5 years). Best of luck! -- With best regards, Grigory Shamov Chemisty, University of Manitoba --- On Tue, 4/28/09, geoffroy.val...@free.fr <geoffroy.val...@free.fr> wrote: > From: geoffroy.val...@free.fr <geoffroy.val...@free.fr> > Subject: Re: [Oscar-users] 6.0.2, 6.0.3 and broken initrd > To: "Ansgar Esztermann" <aesz...@gwdg.de> > Cc: oscar-users@lists.sourceforge.net > Date: Tuesday, April 28, 2009, 8:24 AM > Ansgar, > > Sorry but i cannot help you. You do not give me enough > details, keep continue to say that OSCAR is broken even when > i give you other points to investigate. Sorry i cannot do > more based on the information i have and again, __ALL__ > users how had similar issues (and i know about), the problem > was because the kernel in the image is 2.6.18 based whereas > the SystemImager kernel to image the node is > 2.6.19 > (SATA naming issue and so on -- remember that selecting the > SCSI template when creating the nodes may not be enough), or > because the kernel shipped with RHEL was not working > correctly on their platforms. > > Regards, > > PS: for the kernel difference between the headnode and the > compute nodes, we use online repositories for both of them. > This is not an OSCAR issue, we do not enforce the version. > So most certainly your headnode is not up-to-date or not > using the same repositories, one of them being out-of-date. > > ----- Mail Original ----- > De: "Ansgar Esztermann" <aesz...@gwdg.de> > À: oscar-users@lists.sourceforge.net > Cc: "geoffroy vallee" <geoffroy.val...@free.fr> > Envoyé: Mardi 28 Avril 2009 05h44:16 GMT -05:00 USA/Canada > - États de l'Est > Objet: Re: [Oscar-users] 6.0.2, 6.0.3 and broken initrd > > Hi, > > > On Apr 24, 2009, at 21:22 , geoffroy.val...@free.fr > wrote: > > > Actually this is not the way SystemImager is working. > The /etc/fstab > > is not necessary to prepare and deploy the image. And > actually, i do > > not have any initrd issue with any of the platforms i > am using. > > Moreover, what is the > > OK. Then I probably do not understand where the initrd used > to boot > the deployed nodes comes from. There is one being generated > when oscar > installs the kernel into the image; but since there is no > fstab at > that point, it will not boot. > During the course of the installation, systemconfigurator > is called > from the .master script. However, > /usr/lib/systemconfig/Initrd.pm will > generate an initrd only for kernels that do not already > have one. > > > > problem? When the system reboots, it cannot find the > root file > > system? If so, it is actually not an initrd issue but > > Yes. The first error message after reboot says that > mkrootdev is > expecting a filesystem option (in the initrd, it looks like > this: > mkrootdev -t -o defaults,ro > > Filesystem type and device are missing unless an fstab has > been > present during initrd creation. > Later, the system will panic when trying to switchroot. > > > a problem based on the different between the kernel > issue to image > > the node and then the kernel used to boot the > > OK, maybe kernel versions are somehow mixed up. I > have > kernel-2.6.18-92.el5 in the image; this is not quite the > same as the > kernel on the master (kernel-2.6.18-92.1.22.el5), although > it should > be as far as I understand the docs: > http://svn.oscar.openclustergroup.org/trac/oscar/wiki/SystemImager > > > compute nodes. Please, look at the SystemImager > documentation or > > give us more details about your configuration so we > can try to help > > you. > > I have looked at the wiki at systemimager.org, but I am a > bit lost as > to what to look for. Obviously, the initrd used to boot the > node is > not correct, but it is not clear to me at what point the > initrd should > be generated. Are there any specific parts of the > SystemImager docs I > should take a closer look at? > As to my configuration, that should not be much out of the > ordinary. > CentOS 5.2 (now 5.3), Oscar unstable. The nodes are > SATA-based, so I > made sure to use /usr/share/oscar/oscarsamples/scsi.disk as > a disk > file. It is an x86_64 system. Is there anything else you > would like to > know? > > > PS: i also know for sure that few users solved booting > issues with > > CentOS-5.3 instead of CentOS-5.2, the kernel > configuration is > > slightly different and avoid issues on few platforms. > CentOS-5.3 is > > currently used by default if you are using the > unstable repository. > > Yes, I am using unstable with 5.3. > > > Thank you very much for your help, > > > A. > > > -- > Ansgar Esztermann > DV-Systemadministration > Max-Planck-Institut für biophysikalische Chemie, Abteilung > 105 > > > ------------------------------------------------------------------------------ > Register Now & Save for Velocity, the Web Performance > & Operations > Conference from O'Reilly Media. Velocity features a full > day of > expert-led, hands-on workshops and two days of sessions > from industry > leaders in dedicated Performance & Operations tracks. > Use code vel09scf > and Save an extra 15% before 5/3. http://p.sf.net/sfu/velocityconf > _______________________________________________ > Oscar-users mailing list > Oscar-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/oscar-users > ------------------------------------------------------------------------------ Register Now & Save for Velocity, the Web Performance & Operations Conference from O'Reilly Media. Velocity features a full day of expert-led, hands-on workshops and two days of sessions from industry leaders in dedicated Performance & Operations tracks. Use code vel09scf and Save an extra 15% before 5/3. http://p.sf.net/sfu/velocityconf _______________________________________________ Oscar-users mailing list Oscar-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/oscar-users ------------------------------------------------------------------------------ Register Now & Save for Velocity, the Web Performance & Operations Conference from O'Reilly Media. Velocity features a full day of expert-led, hands-on workshops and two days of sessions from industry leaders in dedicated Performance & Operations tracks. Use code vel09scf and Save an extra 15% before 5/3. http://p.sf.net/sfu/velocityconf _______________________________________________ Oscar-users mailing list Oscar-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/oscar-users
