Re: [Lustre-discuss] Lustre 1.6.5.1 on X4200 and STK 6140 Issues
I know you say the only addition was the RDAC for the MDS's I assume (we use it also just fine). When I ran faultmond from suns dcmu rpm (RHEL 4 here) the x4500's would crash like clock work ~48 hours. For a very simple bit of code I was surpised that once when I forgot to turn it on when working on the load this would happen. Just FYI it was unrelated to lustre (using provided rpm's no kernel build) this solved my problem on the x4500 Brock Palen www.umich.edu/~brockp Center for Advanced Computing [EMAIL PROTECTED] (734)936-1985 On Oct 13, 2008, at 4:41 AM, Malcolm Cowe wrote: The X4200m2 MDS systems and the X4500 OSS were rebuilt using the stock Lustre packages (Kernel + modules + userspace). With the exception of the RDAC kernel module, no additional software was applied to the systems. We recreated our volumes and ran the servers over the weekend. However, the OSS crashed about 8 hours in. The syslog output is attached to this message. Looks like it could be similar to bug #16404, which means patching and rebuilding the kernel. Given my lack of success at trying to build from source, I am again asking for some guidance on how to do this. I sent out the steps I used to try and build from source on the 7th because I was encountering problems and was unable to get a working set of packages. Included in that messages was output from quilt that implies that the kernel patching process was not working properly. Regards, Malcolm. -- 6g_top.gif Malcolm Cowe Solutions Integration Engineer Sun Microsystems, Inc. Blackness Road Linlithgow, West Lothian EH49 7LR UK Phone: x73602 / +44 1506 673 602 Email: [EMAIL PROTECTED] 6g_top.gif Oct 10 06:49:39 oss-1 kernel: LDISKFS FS on md15, internal journal Oct 10 06:49:39 oss-1 kernel: LDISKFS-fs: mounted filesystem with ordered data mode. Oct 10 06:53:42 oss-1 kernel: kjournald starting. Commit interval 5 seconds Oct 10 06:53:42 oss-1 kernel: LDISKFS FS on md16, internal journal Oct 10 06:53:42 oss-1 kernel: LDISKFS-fs: mounted filesystem with ordered data mode. Oct 10 06:57:49 oss-1 kernel: kjournald starting. Commit interval 5 seconds Oct 10 06:57:49 oss-1 kernel: LDISKFS FS on md17, internal journal Oct 10 06:57:49 oss-1 kernel: LDISKFS-fs: mounted filesystem with ordered data mode. Oct 10 07:44:55 oss-1 faultmond: 16:Polling all 48 slots for drive fault Oct 10 07:45:00 oss-1 faultmond: Polling cycle 16 is complete Oct 10 07:56:23 oss-1 kernel: Lustre: OBD class driver, [EMAIL PROTECTED] Oct 10 07:56:23 oss-LDISKFS-fs: file extents enabled1 kernel: Lustre VersionLDISKFS-fs: mballoc enabled : 1.6.5.1 Oct 10 07:56:23 oss-1 kernel: Build Version: 1.6.5.1-1969123119-PRISTINE-.cache.OLDRPMS.20080618230526.linux- smp-2.6.9-67.0.7.EL_lustre.1.6.5.1.x86_64-2.6.9-67.0.7.EL_lustre. 1.6.5.1smp Oct 10 07:56:24 oss-1 kernel: Lustre: Added LNI [EMAIL PROTECTED] [8/64] Oct 10 07:56:24 oss-1 kernel: Lustre: Lustre Client File System; [EMAIL PROTECTED] Oct 10 07:56:24 oss-1 kernel: kjournald starting. Commit interval 5 seconds Oct 10 07:56:24 oss-1 kernel: LDISKFS FS on md11, external journal on md21 Oct 10 07:56:24 oss-1 kernel: LDISKFS-fs: mounted filesystem with journal data mode. Oct 10 07:56:24 oss-1 kernel: kjournald starting. Commit interval 5 seconds Oct 10 07:56:24 oss-1 kernel: LDISKFS FS on md11, external journal on md21 Oct 10 07:56:24 oss-1 kernel: LDISKFS-fs: mounted filesystem with journal data mode. Oct 10 07:56:24 oss-1 kernel: LDISKFS-fs: file extents enabled Oct 10 07:56:24 oss-1 kernel: LDISKFS-fs: mballoc enabled Lustre: Request x1 sent from [EMAIL PROTECTED] to NID [EMAIL PROTECTED] 5s ago has timed out (limit 5s). Oct 10 07:56:30 oss-1 kernel: Lustre: Request x1 sent from [EMAIL PROTECTED] to NID [EMAIL PROTECTED] 5s ago has timed out (limit 5s). LustreError: 4685:0:(events.c:55:request_out_callback()) @@@ type 4, status -113 [EMAIL PROTECTED] x3/t0 o250- [EMAIL PROTECTED]@o2ib_1:26/25 lens 240/400 e 0 to 5 dl 1223621815 ref 2 fl Rpc:/0/0 rc 0/0 Lustre: Request x3 sent from [EMAIL PROTECTED] to NID [EMAIL PROTECTED] 0s ago has timed out (limit 5s). LustreError: 18125:0:(obd_mount.c:1062:server_start_targets()) Required registration failed for lfs01-OST: -5 LustreError: 15f-b: Communication error with the MGS. Is the MGS running? LustreError: 18125:0:(obd_mount.c:1597:server_fill_super()) Unable to start targets: -5 LustreError: 18125:0:(obd_mount.c:1382:server_put_super()) no obd lfs01-OST LustreError: 18125:0:(obd_mount.c:119:server_deregister_mount()) lfs01-OST not registered LDISKFS-fs: mballoc: 0 blocks 0 reqs (0 success) LDISKFS-fs: mballoc: 0 extents scanned, 0 goal hits, 0 2^N hits, 0 breaks, 0 lost LDISKFS-fs: mballoc: 0 generated and it took 0 LDISKFS-fs: mballoc: 0 preallocated, 0 discarded Oct 10 07:56:50 oss-1
Re: [Lustre-discuss] Lustre 1.6.5.1 on X4200 and STK 6140 Issues
I never uninstalled it (i still use some of the tools in it) Faultmond is a service, just chkconfig it off. Brock Palen www.umich.edu/~brockp Center for Advanced Computing [EMAIL PROTECTED] (734)936-1985 On Oct 13, 2008, at 11:03 AM, Malcolm Cowe wrote: Brock Palen wrote: I know you say the only addition was the RDAC for the MDS's I assume (we use it also just fine). Yes, the MDS's share a STK 6140. When I ran faultmond from suns dcmu rpm (RHEL 4 here) the x4500's would crash like clock work ~48 hours. For a very simple bit of code I was surpised that once when I forgot to turn it on when working on the load this would happen. Just FYI it was unrelated to lustre (using provided rpm's no kernel build) this solved my problem on the x4500 The DCMU RPM is installed. I didn't explicitly install this, so it must have been bundled in with the SIA CD... I'll try removing the rpm to see what happens. Thanks for the heads up. Regards, Malcolm. Brock Palen www.umich.edu/~brockp Center for Advanced Computing [EMAIL PROTECTED] (734)936-1985 On Oct 13, 2008, at 4:41 AM, Malcolm Cowe wrote: The X4200m2 MDS systems and the X4500 OSS were rebuilt using the stock Lustre packages (Kernel + modules + userspace). With the exception of the RDAC kernel module, no additional software was applied to the systems. We recreated our volumes and ran the servers over the weekend. However, the OSS crashed about 8 hours in. The syslog output is attached to this message. Looks like it could be similar to bug #16404, which means patching and rebuilding the kernel. Given my lack of success at trying to build from source, I am again asking for some guidance on how to do this. I sent out the steps I used to try and build from source on the 7th because I was encountering problems and was unable to get a working set of packages. Included in that messages was output from quilt that implies that the kernel patching process was not working properly. Regards, Malcolm. -- 6g_top.gif Malcolm Cowe Solutions Integration Engineer Sun Microsystems, Inc. Blackness Road Linlithgow, West Lothian EH49 7LR UK Phone: x73602 / +44 1506 673 602 Email: [EMAIL PROTECTED] 6g_top.gif Oct 10 06:49:39 oss-1 kernel: LDISKFS FS on md15, internal journal Oct 10 06:49:39 oss-1 kernel: LDISKFS-fs: mounted filesystem with ordered data mode. Oct 10 06:53:42 oss-1 kernel: kjournald starting. Commit interval 5 seconds Oct 10 06:53:42 oss-1 kernel: LDISKFS FS on md16, internal journal Oct 10 06:53:42 oss-1 kernel: LDISKFS-fs: mounted filesystem with ordered data mode. Oct 10 06:57:49 oss-1 kernel: kjournald starting. Commit interval 5 seconds Oct 10 06:57:49 oss-1 kernel: LDISKFS FS on md17, internal journal Oct 10 06:57:49 oss-1 kernel: LDISKFS-fs: mounted filesystem with ordered data mode. Oct 10 07:44:55 oss-1 faultmond: 16:Polling all 48 slots for drive fault Oct 10 07:45:00 oss-1 faultmond: Polling cycle 16 is complete Oct 10 07:56:23 oss-1 kernel: Lustre: OBD class driver, [EMAIL PROTECTED] Oct 10 07:56:23 oss-LDISKFS-fs: file extents enabled1 kernel: Lustre VersionLDISKFS-fs: mballoc enabled : 1.6.5.1 Oct 10 07:56:23 oss-1 kernel: Build Version: 1.6.5.1-1969123119-PRISTINE-.cache.OLDRPMS. 20080618230526.linux- smp-2.6.9-67.0.7.EL_lustre. 1.6.5.1.x86_64-2.6.9-67.0.7.EL_lustre. 1.6.5.1smp Oct 10 07:56:24 oss-1 kernel: Lustre: Added LNI [EMAIL PROTECTED] [8/64] Oct 10 07:56:24 oss-1 kernel: Lustre: Lustre Client File System; [EMAIL PROTECTED] Oct 10 07:56:24 oss-1 kernel: kjournald starting. Commit interval 5 seconds Oct 10 07:56:24 oss-1 kernel: LDISKFS FS on md11, external journal on md21 Oct 10 07:56:24 oss-1 kernel: LDISKFS-fs: mounted filesystem with journal data mode. Oct 10 07:56:24 oss-1 kernel: kjournald starting. Commit interval 5 seconds Oct 10 07:56:24 oss-1 kernel: LDISKFS FS on md11, external journal on md21 Oct 10 07:56:24 oss-1 kernel: LDISKFS-fs: mounted filesystem with journal data mode. Oct 10 07:56:24 oss-1 kernel: LDISKFS-fs: file extents enabled Oct 10 07:56:24 oss-1 kernel: LDISKFS-fs: mballoc enabled Lustre: Request x1 sent from [EMAIL PROTECTED] to NID [EMAIL PROTECTED] 5s ago has timed out (limit 5s). Oct 10 07:56:30 oss-1 kernel: Lustre: Request x1 sent from [EMAIL PROTECTED] to NID [EMAIL PROTECTED] 5s ago has timed out (limit 5s). LustreError: 4685:0:(events.c: 55:request_out_callback()) @@@ type 4, status -113 [EMAIL PROTECTED] x3/t0 o250- [EMAIL PROTECTED]@o2ib_1:26/25 lens 240/400 e 0 to 5 dl 1223621815 ref 2 fl Rpc:/0/0 rc 0/0 Lustre: Request x3 sent from [EMAIL PROTECTED] to NID [EMAIL PROTECTED] 0s ago has timed out (limit 5s). LustreError: 18125:0:(obd_mount.c: 1062:server_start_targets()) Required registration failed for lfs01-OST: -5 LustreError: 15f-b: Communication error with the MGS. Is the MGS
Re: [Lustre-discuss] Lustre 1.6.5.1 on X4200 and STK 6140 Issues
On Oct 06, 2008 10:59 -0400, Brian J. Murrell wrote: On Mon, 2008-10-06 at 15:47 +0100, Malcolm Cowe wrote: With respect to the OFED stack used, we are using the latest official software stack supplied by Voltaire. The reason for this is that there is more to OFED than just the kernel modules, including many libraries and tools, None of these should be necessary for Lustre to use I/B. Also very important to note is that if you are changing the IB stack, then Lustre also needs to be recompiled to work with the new IB stack. Cheers, Andreas -- Andreas Dilger Sr. Staff Engineer, Lustre Group Sun Microsystems of Canada, Inc. ___ Lustre-discuss mailing list Lustre-discuss@lists.lustre.org http://lists.lustre.org/mailman/listinfo/lustre-discuss
Re: [Lustre-discuss] Lustre 1.6.5.1 on X4200 and STK 6140 Issues
Andreas Dilger wrote: On Oct 06, 2008 10:59 -0400, Brian J. Murrell wrote: On Mon, 2008-10-06 at 15:47 +0100, Malcolm Cowe wrote: With respect to the OFED stack used, we are using the latest official software stack supplied by Voltaire. The reason for this is that there is more to OFED than just the kernel modules, including many libraries and tools, None of these should be necessary for Lustre to use I/B. Also very important to note is that if you are changing the IB stack, then Lustre also needs to be recompiled to work with the new IB stack. Yes. As a matter of fact, you have anticipated a question I have: how does one re-build Lustre in a safe and consistent manner? I'm working through the docs, but I have come across a problem when I try to run "make rpms" in the Lustre source: make[4]: *** No rule to make target `/usr/src/redhat/BUILD/lustre-1.6.5.1/ldiskfs/Module.symvers', needed by `Module.symvers'. Stop. How do I ensure that the build environment that Lustre requires is properly prepared? I could just hoick a soft link to the Module.symvers file in the kernel tree, but that's a little messy. I've attached a draft copy of the build process to this message. Again, just looking to sanity check the method, since I'm obviously missing something. I'm going to rebuild the servers today so that I can provide the debug information that Brian requested. Cheers, Andreas -- Andreas Dilger Sr. Staff Engineer, Lustre Group Sun Microsystems of Canada, Inc. ___ Lustre-discuss mailing list Lustre-discuss@lists.lustre.org http://lists.lustre.org/mailman/listinfo/lustre-discuss Regards, Malcolm. -- Malcolm Cowe Solutions Integration Engineer Sun Microsystems, Inc. Blackness Road Linlithgow, West Lothian EH49 7LR UK Phone: x73602 / +44 1506 673 602 Email: [EMAIL PROTECTED] ___ Lustre-discuss mailing list Lustre-discuss@lists.lustre.org http://lists.lustre.org/mailman/listinfo/lustre-discuss
Re: [Lustre-discuss] Lustre 1.6.5.1 on X4200 and STK 6140 Issues
On Oct 06, 2008 10:24 -0400, Ms. Megan Larko wrote: The order I used which generated no unknown symbol errors for installation of lustre 1.6.5.1 was this: 1) kernel-lustre-smp-2.6.18-53.1.14.el5_lustre.1.6.5.1.x86_64.rpm If using infiniband (IB) this is next: 2) kernel-ib-1.3-2.6.18_53.1.14.el5_lustre.1.6.5.1smp.x86_64.rpm 3) lustre-ldiskfs-3.0.4-2.6.18_53.1.14.el5_lustre.1.6.5.1smp.x86_64.rpm 4) lustre-modules-1.6.5.1-2.6.18_53.1.14.el5_lustre.1.6.5.1smp.x86_64.rpm 5) lustre-1.6.5.1-2.6.18_53.1.14.el5_lustre.1.6.5.1smp.x86_64.rpm That is good to know for the documentation. However, I suspect if all of these packages are installed at the same time there would also not be any symbol warnings. If a module installation does have many unknown symbol references, then find the rpm which will satisfy those references and install that module. To actually have them satisfied one must return to the package that had complained about the unknown symbol and having already installed the package to satisfy those symbols then rpm --force -ivh to force a retry of the package with the issues. That isn't quite correct. The missing module symbols are the output of depmod -ae that is run in the RPM post-install after kernel modules are installed. Even if there are such warnings, if the modules are later installed and depmod -ae is run again it should report no warnings, regardless of what order the modules were installed. That means - no need to reinstall the RPMs or to install them in a particular order, though of course avoid the warnings is always nicer. You can always run depmod -ae by hand to re-verify the modules of the currently installed kernels. Cheers, Andreas -- Andreas Dilger Sr. Staff Engineer, Lustre Group Sun Microsystems of Canada, Inc. ___ Lustre-discuss mailing list Lustre-discuss@lists.lustre.org http://lists.lustre.org/mailman/listinfo/lustre-discuss
Re: [Lustre-discuss] Lustre 1.6.5.1 on X4200 and STK 6140 Issues
Now with attachment. Sorry. Malcolm Cowe wrote: Andreas Dilger wrote: On Oct 06, 2008 10:59 -0400, Brian J. Murrell wrote: On Mon, 2008-10-06 at 15:47 +0100, Malcolm Cowe wrote: With respect to the OFED stack used, we are using the latest official software stack supplied by Voltaire. The reason for this is that there is more to OFED than just the kernel modules, including many libraries and tools, None of these should be necessary for Lustre to use I/B. Also very important to note is that if you are changing the IB stack, then Lustre also needs to be recompiled to work with the new IB stack. Yes. As a matter of fact, you have anticipated a question I have: how does one re-build Lustre in a safe and consistent manner? I'm working through the docs, but I have come across a problem when I try to run "make rpms" in the Lustre source: make[4]: *** No rule to make target `/usr/src/redhat/BUILD/lustre-1.6.5.1/ldiskfs/Module.symvers', needed by `Module.symvers'. Stop. How do I ensure that the build environment that Lustre requires is properly prepared? I could just hoick a soft link to the Module.symvers file in the kernel tree, but that's a little messy. I've attached a draft copy of the build process to this message. Again, just looking to sanity check the method, since I'm obviously missing something. I'm going to rebuild the servers today so that I can provide the debug information that Brian requested. Cheers, Andreas -- Andreas Dilger Sr. Staff Engineer, Lustre Group Sun Microsystems of Canada, Inc. ___ Lustre-discuss mailing list Lustre-discuss@lists.lustre.org http://lists.lustre.org/mailman/listinfo/lustre-discuss Regards, Malcolm. -- Malcolm Cowe Solutions Integration Engineer Sun Microsystems, Inc. Blackness Road Linlithgow, West Lothian EH49 7LR UK Phone: x73602 / +44 1506 673 602 Email: [EMAIL PROTECTED] -- Malcolm Cowe Solutions Integration Engineer Sun Microsystems, Inc. Blackness Road Linlithgow, West Lothian EH49 7LR UK Phone: x73602 / +44 1506 673 602 Email: [EMAIL PROTECTED] Creating Lustre Packages From Source for RHEL 4.5 AS, 64-bit Pre-requisites -- RHEL 4.5 AS Full Installation SUN-supplied Linux RDAC kernel modules installed. System running on Stock RHEL 4.5 AS kernel with RDAC support. Quilt Source Lustre Source RHEL 4.5 AS Kernel SRPM 1. Download and install Quilt: http://download.savannah.gnu.org/releases/quilt/ 2. Install the RHEL 4.5 AS Kernel SRPM, found on RHEL 4.5 source CD 4. 3. Change to the Red Hat RPM specs directory and extract the full kernel source tree. This will also apply Red Hat's patches to the source: cd /usr/src/redhat/SPECS rpmbuild -bp kernel-2.6.spec 4. Install the Lustre sources. rpm -ivh lustre-source-1.6.5.1-2.6.9_67.0.7.EL_lustre.1.6.5.1smp.x86_64.rpm 5. Prepare the kernel source for the Lustre patches: cd /usr/src/redhat/BUILD/kernel-2.6.9/linux-2.6.9 rm -f patches series ln -s /usr/src/lustre-1.6.5.1/lustre/kernel_patches/series/2.6-rhel4.series series ln -s /usr/src/lustre-1.6.5.1/lustre/kernel_patches/patches . 6. Apply the Lustre patches to the kernel sources: cd /usr/src/redhat/BUILD/kernel-2.6.9/linux-2.6.9 quilt push -av Points to note in output: Hunk #1 FAILED at 770. 1 out of 1 hunk FAILED -- rejects in file fs/nfs/nfs4proc.c Patch patches/vfs_intent-2.6-rhel4.patch does not apply (enforce with -f) Full output: [EMAIL PROTECTED] linux-2.6.9]# quilt push -av Applying patch patches/vfs_intent-2.6-rhel4.patch patching file fs/cifs/dir.c patching file fs/exec.c patching file fs/inode.c patching file fs/namei.c patching file fs/namespace.c patching file fs/nfs/dir.c patching file fs/nfs/nfs4proc.c Hunk #1 FAILED at 770. 1 out of 1 hunk FAILED -- rejects in file fs/nfs/nfs4proc.c patching file fs/open.c patching file fs/stat.c patching file include/linux/dcache.h patching file include/linux/fs.h patching file include/linux/mount.h patching file include/linux/namei.h Restoring include/linux/dcache.h Restoring include/linux/namei.h Restoring include/linux/mount.h Restoring include/linux/fs.h Restoring fs/stat.c Restoring fs/open.c Restoring fs/namespace.c Restoring fs/exec.c Restoring fs/nfs/nfs4proc.c Restoring fs/nfs/dir.c Restoring fs/namei.c Restoring fs/cifs/dir.c Restoring fs/inode.c Patch patches/vfs_intent-2.6-rhel4.patch does not apply (enforce with -f) Restoring include/linux/dcache.h Restoring include/linux/namei.h Restoring include/linux/mount.h Restoring include/linux/fs.h Restoring
[Lustre-discuss] Lustre 1.6.5.1 on X4200 and STK 6140 Issues
Hi Folks, We are trying to create a small lustre environment on behalf of a customer. There are 2 X4200m2 MDS servers, both dual-attached to an STK 6140 array over FC. This is an active-passive arrangement with a single shared volume. Heartbeat is used to co-ordinate file system failover. There is a single X4500 OSS server, the storage for which is split into 6 OSTs. Finally, we have 2 X4600m2 clients, just for kicks. All systems are connected together over ethernet and infiniband, with the IB network being used for Lustre and every system is running RHEL 4.5 AS. The X4500 OST volumes are created using software RAID, while the X4200m2 MDT is accessed using DM Multipath. We downloaded the Lustre binary packages from SUN's web site and installed them onto each of the servers. Unfortunately, the resulting system is very unstable and is prone to lock-ups on the servers (uptimes are measured in hours). These lock-ups happen without warning, and with very little, if any, debug information in the system logs. We have also observed the servers locking up on shutdown (kernel panics). Based on the documentation in the Lustre operations manual, we installed the RPMs as follows: rpm -Uvh --force e2fsprogs-1.40.7.sun3-0redhat.x86_64.rpm rpm -ivh kernel-lustre-smp-2.6.9-67.0.7.EL_lustre.1.6.5.1.x86_64.rpm rpm -ivh kernel-lustre-source-2.6.9-67.0.7.EL_lustre.1.6.5.1.x86_64.rpm rpm -ivh lustre-modules-1.6.5.1-2.6.9_67.0.7.EL_lustre.1.6.5.1smp.x86_64.rpm # (many "unknown symbol" warnings) rpm -ivh lustre-1.6.5.1-2.6.9_67.0.7.EL_lustre.1.6.5.1smp.x86_64.rpm rpm -ivh lustre-source-1.6.5.1-2.6.9_67.0.7.EL_lustre.1.6.5.1smp.x86_64.rpm rpm -ivh lustre-ldiskfs-3.0.4-2.6.9_67.0.7.EL_lustre.1.6.5.1smp.x86_64.rpm # (many "unknown symbol" warnings) mv /etc/init.d/openibd /etc/init.d/openibd.rhel4default rpm -ivh --force kernel-ib-1.3-2.6.9_67.0.7.EL_lustre.1.6.5.1smp.x86_64.rpm cp /etc/init.d/openibd /etc/init.d/openibd.lustre.1.6.5.1 We then reboot the system and load RHEL using the Lustre kernel. Now we install the Voltaire OFED software: Copy the kernel config used to build the Lustre patched kernel into the Lustre kernel source tree: cp /boot/config-2.6.9-67.0.7.EL_lustre.1.6.5.1smp \ /usr/src/linux-2.6.9-67.0.7.EL_lustre.1.6.5.1/.config Change into the Lustre kernel source and edit the Makefile. Change "custom" suffix to "smp" in the variable "EXTRAVERSION". Change into the lustre kernel source and run these setup commands: make oldconfig || make menuconfig make include/asm make include/linux/version.h make SUBDIRS=scripts Change into the "-obj" directory and run these setup commands: cd /usr/src/linux-2.6.9-67.0.7.EL_lustre.1.6.5.1-obj/x86_64/smp ln -s /usr/src/linux-2.6.9-67.0.7.EL_lustre.1.6.5.1/include . Unpack the Voltaire OFED tar-ball: tar zxf VoltaireOFED-5.1.3.1_5.tgz Change to the unpacked software directory and run the installation script. To build the OFED packages with the Voltaire certified configuration, run the following commands: cd VoltaireOFED-5.1.3.1_5 ./install.pl -c ofed.conf.Volt Once complete, reboot. Configure any IPoIB interfaces as required. Add the following into /etc/modprobe.conf: options lnet networks="o2ib0(ib0)" Load the Lustre LNET kernel module. modprobe lnet Start the Lustre core networking service. lctl network up Check the system log (/var/log/messages) for confirmation. Create the MGS/MDT Lustre Volume: Format the MGS/MDT device. mkfs.lustre [ --reformat ] --fsname lfs01 --mdt --mgs [EMAIL PROTECTED] /dev/dm-0 Create the MGS/MDT file system mount point. mkdir -p /lustre/mdt/lfs01 Mount the file system. This will initiate MGS and MDT services for Lustre. mount -t lustre /dev/dm-0 /lustre/mdt/lfs01 With the exception of the OST volume creation, we use an equivalent process to bring the OSS online. The cabling has been checked and verified. So we re-built the system from scratch and applied only SUN's RDAC modules and Voltaire OFED to the stock RHEL 4.5 kernel (2.6.9-55.ELsmp). We removed the second MDS from the h/w configuration and did not install Heartbeat. The shared storage was re-formatted as a regular EXT3 file system using the DM multipathing device, /dev/dm-0, and mounted onto the host. Running I/O tests onto the mounted file system over an extended period did not elicit a single error or warning message in the log related to the multipathing or the SCSI device. Once we were confident that the system was running in a consistent and stable manner, we re-installed the Lustre packages, omitting the kernel-ib packages. We had to re-build and re-install the RDAC support as well. This means that the system has support for the Lustre file system but no infiniband support at all. /etc/modprobe.conf is updated such that the lnet networks option is set to "tcp".
[Lustre-discuss] Lustre 1.6.5.1 on X4200 and STK 6140 Issues
Hello, Reading through the message from Malcolm Cowe about a new lustre environment, he mentioned that there were unknown symbol warnings during his installation procedure. Well, I also saw warnings when I was doing a lustre 1.6.5.1 install. I know from general linux that the rpm is not properly installed while there are so many warning and of the type (some were ldiskfs issues for example) that I was seeing. What I discovered is that the order in which the lustre 1.6.5.1 rpms are installed does matter and that it is not the same order as indicated in the Lustre Manual version 1.12 for 1.6.4. The order I used which generated no unknown symbol errors for installation of lustre 1.6.5.1 was this: 1) kernel-lustre-smp-2.6.18-53.1.14.el5_lustre.1.6.5.1.x86_64.rpm If using infiniband (IB) this is next: 2) kernel-ib-1.3-2.6.18_53.1.14.el5_lustre.1.6.5.1smp.x86_64.rpm 3) lustre-ldiskfs-3.0.4-2.6.18_53.1.14.el5_lustre.1.6.5.1smp.x86_64.rpm 4) lustre-modules-1.6.5.1-2.6.18_53.1.14.el5_lustre.1.6.5.1smp.x86_64.rpm 5) lustre-1.6.5.1-2.6.18_53.1.14.el5_lustre.1.6.5.1smp.x86_64.rpm The above were done using rpm -i (install) which works well for kernels so that you have multiple versions (hopefully including a good one to which one may return if necessary). The last cannot be done using -i, but -U: rpm -Uvh e2fsprogs-1.40.7.sun3-0redhat.x86_64.rpm This has not been a problem for us as the newer version seems to get along fine with a 1.6.4.3 version of Lustre (I appreciate backwards compatibility). If a module installation does have many unknown symbol references, then find the rpm which will satisfy those references and install that module. To actually have them satisfied one must return to the package that had complained about the unknown symbol and having already installed the package to satisfy those symbols then rpm --force -ivh to force a retry of the package with the issues. This procedure can be iterative as sometimes more than one package may be needed to satisfy all the references of the package desired. I do know from personal experience that if a package has unknown symbols especially if those symbols are used/accessed, then it can panic the box. My experience with this in on 64-bit hw using Cent OS 5 as the base Operating System. Best of luck. megan ___ Lustre-discuss mailing list Lustre-discuss@lists.lustre.org http://lists.lustre.org/mailman/listinfo/lustre-discuss
Re: [Lustre-discuss] Lustre 1.6.5.1 on X4200 and STK 6140 Issues
Hey Brian, I'll have to re-install the system from scratch in order to be able to answer some of your questions, which I'll get started on this evening. What I was hoping for in the first instance was a sanity check of our installation methods. With respect to the OFED stack used, we are using the latest official software stack supplied by Voltaire. The reason for this is that there is more to OFED than just the kernel modules, including many libraries and tools, plus the latest firmware for the cards. It's what the customer has asked for, and it is what the card vendor expects us to do. We may be able to get away with OFED 1.3, but I would still like some guidance on how to install the rest of the OFED stack -- do we use the OFED source to rebuild everything, or can we pick the Lustre supplied kernel modules and just layer on the other stuff separately? Like I said, sanity-checking the install procedure is important. Finally, when I said that one file system fails versus another passes, I mean that the server locks solid, crashes, usually with no debug to speak of (nothing in the system logs). Even while the system is up and running the lustre kernel, if we attempt a clean shutdown, the kernel panics. Since I need to rebuild the systems anyway, I will also try to install the packages in the order mentioned by Megan Larko, to see how that affects the installation. We have been following the instructions in the Lustre Operations Manual (v. 1.14). Regards, Malcolm. Brian J. Murrell wrote: On Mon, 2008-10-06 at 10:58 +0100, Malcolm Cowe wrote: rpm -Uvh --force e2fsprogs-1.40.7.sun3-0redhat.x86_64.rpm You should not (have to) use --force. If you do, there is either an operational error or a bug in our packages. In the latter case, please file a bug in our bugzilla. rpm -ivh lustre-modules-1.6.5.1-2.6.9_67.0.7.EL_lustre.1.6.5.1smp.x86_64.rpm # (many "unknown symbol" warnings) Can you paste them here? rpm -ivh lustre-ldiskfs-3.0.4-2.6.9_67.0.7.EL_lustre.1.6.5.1smp.x86_64.rpm # (many "unknown symbol" warnings) Ditto. rpm -ivh --force kernel-ib-1.3-2.6.9_67.0.7.EL_lustre.1.6.5.1smp.x86_64.rpm Again, you should not need to use --force. We then reboot the system and load RHEL using the Lustre kernel. Now we install the Voltaire OFED software: Why? The kernel-ib package you installed above should provide a working OFED stack. 1. Unpack the Voltaire OFED tar-ball: tar zxf VoltaireOFED-5.1.3.1_5.tgz Do you really need 1.3.1? If so, then you should not install the 1.3 kernel-ib package we provide above. I really wonder why you need 1.3.1 though. * Lustre supplied kernel, Lustre software. No IB. MDS/MGS file system. FAILED. Failed in what way? * Lustre supplied kernel, Lustre software, RDAC. No IB. MDS/MGS file system (Full Lustre FS over Ethernet). FAILED. Again, in what way? * Lustre supplied kernel, Lustre software, RDAC, Voltaire OFED. EXT-3 file system. FAILED. Ditto. * Lustre supplied kernel, Lustre software. RDAC, Voltaire OFED. MDS/MGS file system (Full Lustre FS over IB). FAILED. And Ditto again. You have to provide more details than just "FAILED" if we are to try to help diagnose a problem. Our findings indicate that there is a problem within the binary distribution of Lustre. I think that many of our users use it as is, so it cannot be all that bad. This may be due to the fact that we are applying the 2.6.9-67 RHEL kernel to a platform based upon 2.6.9.-55, That shouldn't be a problem in and of itself. b. ___ Lustre-discuss mailing list Lustre-discuss@lists.lustre.org http://lists.lustre.org/mailman/listinfo/lustre-discuss -- Malcolm Cowe Solutions Integration Engineer Sun Microsystems, Inc. Blackness Road Linlithgow, West Lothian EH49 7LR UK Phone: x73602 / +44 1506 673 602 Email: [EMAIL PROTECTED] ___ Lustre-discuss mailing list Lustre-discuss@lists.lustre.org http://lists.lustre.org/mailman/listinfo/lustre-discuss
Re: [Lustre-discuss] Lustre 1.6.5.1 on X4200 and STK 6140 Issues
On Mon, 2008-10-06 at 15:47 +0100, Malcolm Cowe wrote: Hey Brian, Hey Malcolm, I'll have to re-install the system from scratch in order to be able to answer some of your questions, which I'll get started on this evening. OK. What I was hoping for in the first instance was a sanity check of our installation methods. I think I commented on those. If you are going to build your OFED stack you don't need to install the one we provide. With respect to the OFED stack used, we are using the latest official software stack supplied by Voltaire. The reason for this is that there is more to OFED than just the kernel modules, including many libraries and tools, None of these should be necessary for Lustre to use I/B. plus the latest firmware for the cards. Hrm. Can you not upgrade firmware independent of upgrading the whole OFED stack? That seems very limiting. It's what the customer has asked for, and it is what the card vendor expects us to do. Fair enough. I was just pointing out that you don't need our OFED stack if you are going to install your own. We may be able to get away with OFED 1.3, but I would still like some guidance on how to install the rest of the OFED stack We don't supply the userspace tools because they are not really necessary for Lustre. do we use the OFED source to rebuild everything, or can we pick the Lustre supplied kernel modules and just layer on the other stuff separately? Yes, you should be able to do that. I say that quite generally as I'm not entirely clear on your operating environment. Finally, when I said that one file system fails versus another passes, I mean that the server locks solid, crashes, usually with no debug to speak of (nothing in the system logs). Nothing on the console either? Even while the system is up and running the lustre kernel, if we attempt a clean shutdown, the kernel panics. Hrm. A panic is quite different than locking solid with no messages at all. A solid lock with no messages is indicative of hardware problems. Since I need to rebuild the systems anyway, I will also try to install the packages in the order mentioned by Megan Larko, to see how that affects the installation. I'm not entirely convinced of her process. You should not need to use --force and reinstall packages already installed. I'd be more interested in knowing exactly your installation steps and the errors you get from it. Please try to avoid the use of --force so we can see why it's necessary. You will have to use rpm -U with e2fsprogs though as she mentions. Do all of your work with the script(1) tool so you can easily log it. b. signature.asc Description: This is a digitally signed message part ___ Lustre-discuss mailing list Lustre-discuss@lists.lustre.org http://lists.lustre.org/mailman/listinfo/lustre-discuss