[lustre-discuss] Building on Centos 7.4
Hi List! I'm running into a bunch of trouble building Lustre 2.10.2 on CentOS 7.4.1708 (3.10.0-693) and I'm hoping that someone here can help. Building SPL, ZFS, and Lustre all complete without errors and I'm left with a bunch of RPMs that appear to be complete. However, when I go to install Lustre yum reports this, and many similar, errors: kmod-lustre-osd-zfs-2.10.2-1.el7.centos.x86_64 Requires: ksym(zap_cursor_retrieve) =0x3f9d6ed5 I gather this has something to do with not being able to find the correct symbols for ZFS but I'm not sure where I'm failing. My build an installation procedure follows. Any advice would be greatly appreciated! First, I built and installed SPL and ZFS by downloading the 0.7.3 release tarballs and performing the following: cd ./spl-0.7.3 ./configure make pkg-utils pkg-kmod yum localinstall ./*.rpm cd ../zfs-0.7.3 ./configure make pkg-utils pkg-kmod yum localinstall ./*.rpm Next, I build and try to install Lustre: cd ../lustre-release-f3f2a2f bash ./autogen.sh ./configure --with-o2ib=yes --enable-server make rpms yum localinstall ./*.rpm The configure script and make both report no errors and the zfs and spl modules appear to be loaded. Any thoughts? -- Neale Petrillo ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
Re: [lustre-discuss] lnet shutdown issue while cleint reboot
Please run "lctl get_param version" on a client to find the version currently running. There is a problem if you don't have libyaml-devel installed at build time then lnetctl will not be built. As of 2.10.2 the libyaml-devel library is required so that lnetctl is always built. Cheers, Andreas > On Jan 3, 2018, at 08:41, Parag Khuraswarwrote: > > Hi Jones, > > I installed this version 2 months back. But now I am facing this client > reboot issue. Lustre has been setup and working properly. > > Regards, > Parag > > > > -Original Message- > From: Jones, Peter A [mailto:peter.a.jo...@intel.com] > Sent: Wednesday, January , 2018 8:31 PM > To: Parag Khuraswar; 'Arman Khalatyan' > Cc: 'Lustre discussion' > Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot > > Ok, so that is the correct place and you meant git rather than GitHub. So > what tag are you building against? As mentioned previously 2.20 is not a > version of Lustre so perhaps this is a typo too? > > > > >> On 2018-01-03, 6:54 AM, "lustre-discuss on behalf of Parag Khuraswar" >> > para...@citilindia.com> wrote: >> >> Hi Jones, >> >> I cloned from here- >> >> git clone git://git.hpdd.intel.com/fs/lustre-release.git >> >> Regards, >> Parag >> >> >> >> -Original Message- >> From: Jones, Peter A [mailto:peter.a.jo...@intel.com] >> Sent: Wednesday, January , 2018 7:52 PM >> To: Parag Khuraswar; 'Arman Khalatyan' >> Cc: 'Lustre discussion' >> Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot >> >> What location on GitHub? Do you mean the IML repo? >> >> >> >> >>> On 2018-01-03, 6:04 AM, "lustre-discuss on behalf of Parag Khuraswar" >>> >> para...@citilindia.com> wrote: >>> >>> I cloned from github. Lnetctl is there on lustre servers but on clients >>> only lctl is available. >>> >>> Regards, >>> Parag >>> >>> >>> >>> -Original Message- >>> From: Arman Khalatyan [mailto:arm2...@gmail.com] >>> Sent: Wednesday, January , 2018 7:06 PM >>> To: Parag Khuraswar >>> Cc: Lustre discussion >>> Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot >>> >>> Strange, Is it some custom lustre ? the 2.20 is not yet there: >>> http://lustre.org/download/ >>> lnetctl is inside since 2.10.x >>> >>> On Wed, Jan 3, 2018 at 12:46 PM, Parag Khuraswar wrote: Hi, In my version of lustre on client nodes 'lnetctl " command is not available. Regards, Parag -Original Message- From: Arman Khalatyan [mailto:arm2...@gmail.com] Sent: Wednesday, January , 2018 4:54 PM To: Parag Khuraswar Cc: Lustre discussion Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot hi, Try this before reboot: umount /lustre service lnet stop lnetctl lnet unconfigure lustre_rmmod then reboot On Centos 7.4 it works. Cheers, Arman. > On Wed, Jan 3, 2018 at 10:39 AM, Parag Khuraswar > wrote: > Hi, > > > > I am using lustre 2.20.1 on RHEL 7.3. On the lustre client nodes > when I shutdown I get attached error and nodes don’t get shutdown. > > The procedure I follow to shut down the nodes are- > > 1) Unmounts all lustre file systems, > > 2) Stop lnet service ( lnet service stops successfully.), > > 3) Unload lustre module, > > > > After performing above steps and rebooting node it sucks, error is > attached. > > > > Regards, > > Parag > > > > > > > ___ > lustre-discuss mailing list > lustre-discuss@lists.lustre.org > http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org > >>> >>> ___ >>> lustre-discuss mailing list >>> lustre-discuss@lists.lustre.org >>> http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org >> >> ___ >> lustre-discuss mailing list >> lustre-discuss@lists.lustre.org >> http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org > > ___ > lustre-discuss mailing list > lustre-discuss@lists.lustre.org > http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
Re: [lustre-discuss] lnet shutdown issue while cleint reboot
Hi Jones, I installed this version 2 months back. But now I am facing this client reboot issue. Lustre has been setup and working properly. Regards, Parag -Original Message- From: Jones, Peter A [mailto:peter.a.jo...@intel.com] Sent: Wednesday, January , 2018 8:31 PM To: Parag Khuraswar; 'Arman Khalatyan' Cc: 'Lustre discussion' Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot Ok, so that is the correct place and you meant git rather than GitHub. So what tag are you building against? As mentioned previously 2.20 is not a version of Lustre so perhaps this is a typo too? On 2018-01-03, 6:54 AM, "lustre-discuss on behalf of Parag Khuraswar"wrote: >Hi Jones, > >I cloned from here- > >git clone git://git.hpdd.intel.com/fs/lustre-release.git > >Regards, >Parag > > > >-Original Message- >From: Jones, Peter A [mailto:peter.a.jo...@intel.com] >Sent: Wednesday, January , 2018 7:52 PM >To: Parag Khuraswar; 'Arman Khalatyan' >Cc: 'Lustre discussion' >Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot > >What location on GitHub? Do you mean the IML repo? > > > > >On 2018-01-03, 6:04 AM, "lustre-discuss on behalf of Parag Khuraswar" > >wrote: > >>I cloned from github. Lnetctl is there on lustre servers but on clients only >>lctl is available. >> >>Regards, >>Parag >> >> >> >>-Original Message- >>From: Arman Khalatyan [mailto:arm2...@gmail.com] >>Sent: Wednesday, January , 2018 7:06 PM >>To: Parag Khuraswar >>Cc: Lustre discussion >>Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot >> >>Strange, Is it some custom lustre ? the 2.20 is not yet there: >>http://lustre.org/download/ >>lnetctl is inside since 2.10.x >> >> >>On Wed, Jan 3, 2018 at 12:46 PM, Parag Khuraswar >>wrote: >>> Hi, >>> >>> In my version of lustre on client nodes 'lnetctl " command is not available. >>> >>> Regards, >>> Parag >>> >>> >>> >>> -Original Message- >>> From: Arman Khalatyan [mailto:arm2...@gmail.com] >>> Sent: Wednesday, January , 2018 4:54 PM >>> To: Parag Khuraswar >>> Cc: Lustre discussion >>> Subject: Re: [lustre-discuss] lnet shutdown issue while cleint >>> reboot >>> >>> hi, >>> Try this before reboot: >>> umount /lustre >>> service lnet stop >>> lnetctl lnet unconfigure >>> lustre_rmmod >>> then reboot >>> On Centos 7.4 it works. >>> >>> Cheers, >>> Arman. >>> >>> >>> On Wed, Jan 3, 2018 at 10:39 AM, Parag Khuraswar >>> wrote: Hi, I am using lustre 2.20.1 on RHEL 7.3. On the lustre client nodes when I shutdown I get attached error and nodes don’t get shutdown. The procedure I follow to shut down the nodes are- 1) Unmounts all lustre file systems, 2) Stop lnet service ( lnet service stops successfully.), 3) Unload lustre module, After performing above steps and rebooting node it sucks, error is attached. Regards, Parag ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org >>> >> >>___ >>lustre-discuss mailing list >>lustre-discuss@lists.lustre.org >>http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org > >___ >lustre-discuss mailing list >lustre-discuss@lists.lustre.org >http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
Re: [lustre-discuss] lnet shutdown issue while cleint reboot
Ok, so that is the correct place and you meant git rather than GitHub. So what tag are you building against? As mentioned previously 2.20 is not a version of Lustre so perhaps this is a typo too? On 2018-01-03, 6:54 AM, "lustre-discuss on behalf of Parag Khuraswar"wrote: >Hi Jones, > >I cloned from here- > >git clone git://git.hpdd.intel.com/fs/lustre-release.git > >Regards, >Parag > > > >-Original Message- >From: Jones, Peter A [mailto:peter.a.jo...@intel.com] >Sent: Wednesday, January , 2018 7:52 PM >To: Parag Khuraswar; 'Arman Khalatyan' >Cc: 'Lustre discussion' >Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot > >What location on GitHub? Do you mean the IML repo? > > > > >On 2018-01-03, 6:04 AM, "lustre-discuss on behalf of Parag Khuraswar" > >wrote: > >>I cloned from github. Lnetctl is there on lustre servers but on clients only >>lctl is available. >> >>Regards, >>Parag >> >> >> >>-Original Message- >>From: Arman Khalatyan [mailto:arm2...@gmail.com] >>Sent: Wednesday, January , 2018 7:06 PM >>To: Parag Khuraswar >>Cc: Lustre discussion >>Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot >> >>Strange, Is it some custom lustre ? the 2.20 is not yet there: >>http://lustre.org/download/ >>lnetctl is inside since 2.10.x >> >> >>On Wed, Jan 3, 2018 at 12:46 PM, Parag Khuraswar >>wrote: >>> Hi, >>> >>> In my version of lustre on client nodes 'lnetctl " command is not available. >>> >>> Regards, >>> Parag >>> >>> >>> >>> -Original Message- >>> From: Arman Khalatyan [mailto:arm2...@gmail.com] >>> Sent: Wednesday, January , 2018 4:54 PM >>> To: Parag Khuraswar >>> Cc: Lustre discussion >>> Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot >>> >>> hi, >>> Try this before reboot: >>> umount /lustre >>> service lnet stop >>> lnetctl lnet unconfigure >>> lustre_rmmod >>> then reboot >>> On Centos 7.4 it works. >>> >>> Cheers, >>> Arman. >>> >>> >>> On Wed, Jan 3, 2018 at 10:39 AM, Parag Khuraswar >>> wrote: Hi, I am using lustre 2.20.1 on RHEL 7.3. On the lustre client nodes when I shutdown I get attached error and nodes don’t get shutdown. The procedure I follow to shut down the nodes are- 1) Unmounts all lustre file systems, 2) Stop lnet service ( lnet service stops successfully.), 3) Unload lustre module, After performing above steps and rebooting node it sucks, error is attached. Regards, Parag ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org >>> >> >>___ >>lustre-discuss mailing list >>lustre-discuss@lists.lustre.org >>http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org > >___ >lustre-discuss mailing list >lustre-discuss@lists.lustre.org >http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
Re: [lustre-discuss] lnet shutdown issue while cleint reboot
Hi Jones, I cloned from here- git clone git://git.hpdd.intel.com/fs/lustre-release.git Regards, Parag -Original Message- From: Jones, Peter A [mailto:peter.a.jo...@intel.com] Sent: Wednesday, January , 2018 7:52 PM To: Parag Khuraswar; 'Arman Khalatyan' Cc: 'Lustre discussion' Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot What location on GitHub? Do you mean the IML repo? On 2018-01-03, 6:04 AM, "lustre-discuss on behalf of Parag Khuraswar"wrote: >I cloned from github. Lnetctl is there on lustre servers but on clients only >lctl is available. > >Regards, >Parag > > > >-Original Message- >From: Arman Khalatyan [mailto:arm2...@gmail.com] >Sent: Wednesday, January , 2018 7:06 PM >To: Parag Khuraswar >Cc: Lustre discussion >Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot > >Strange, Is it some custom lustre ? the 2.20 is not yet there: >http://lustre.org/download/ >lnetctl is inside since 2.10.x > > >On Wed, Jan 3, 2018 at 12:46 PM, Parag Khuraswar >wrote: >> Hi, >> >> In my version of lustre on client nodes 'lnetctl " command is not available. >> >> Regards, >> Parag >> >> >> >> -Original Message- >> From: Arman Khalatyan [mailto:arm2...@gmail.com] >> Sent: Wednesday, January , 2018 4:54 PM >> To: Parag Khuraswar >> Cc: Lustre discussion >> Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot >> >> hi, >> Try this before reboot: >> umount /lustre >> service lnet stop >> lnetctl lnet unconfigure >> lustre_rmmod >> then reboot >> On Centos 7.4 it works. >> >> Cheers, >> Arman. >> >> >> On Wed, Jan 3, 2018 at 10:39 AM, Parag Khuraswar >> wrote: >>> Hi, >>> >>> >>> >>> I am using lustre 2.20.1 on RHEL 7.3. On the lustre client nodes >>> when I shutdown I get attached error and nodes don’t get shutdown. >>> >>> The procedure I follow to shut down the nodes are- >>> >>> 1) Unmounts all lustre file systems, >>> >>> 2) Stop lnet service ( lnet service stops successfully.), >>> >>> 3) Unload lustre module, >>> >>> >>> >>> After performing above steps and rebooting node it sucks, error is attached. >>> >>> >>> >>> Regards, >>> >>> Parag >>> >>> >>> >>> >>> >>> >>> ___ >>> lustre-discuss mailing list >>> lustre-discuss@lists.lustre.org >>> http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org >>> >> > >___ >lustre-discuss mailing list >lustre-discuss@lists.lustre.org >http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
Re: [lustre-discuss] lnet shutdown issue while cleint reboot
What location on GitHub? Do you mean the IML repo? On 2018-01-03, 6:04 AM, "lustre-discuss on behalf of Parag Khuraswar"wrote: >I cloned from github. Lnetctl is there on lustre servers but on clients only >lctl is available. > >Regards, >Parag > > > >-Original Message- >From: Arman Khalatyan [mailto:arm2...@gmail.com] >Sent: Wednesday, January , 2018 7:06 PM >To: Parag Khuraswar >Cc: Lustre discussion >Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot > >Strange, Is it some custom lustre ? the 2.20 is not yet there: >http://lustre.org/download/ >lnetctl is inside since 2.10.x > > >On Wed, Jan 3, 2018 at 12:46 PM, Parag Khuraswar >wrote: >> Hi, >> >> In my version of lustre on client nodes 'lnetctl " command is not available. >> >> Regards, >> Parag >> >> >> >> -Original Message- >> From: Arman Khalatyan [mailto:arm2...@gmail.com] >> Sent: Wednesday, January , 2018 4:54 PM >> To: Parag Khuraswar >> Cc: Lustre discussion >> Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot >> >> hi, >> Try this before reboot: >> umount /lustre >> service lnet stop >> lnetctl lnet unconfigure >> lustre_rmmod >> then reboot >> On Centos 7.4 it works. >> >> Cheers, >> Arman. >> >> >> On Wed, Jan 3, 2018 at 10:39 AM, Parag Khuraswar >> wrote: >>> Hi, >>> >>> >>> >>> I am using lustre 2.20.1 on RHEL 7.3. On the lustre client nodes when >>> I shutdown I get attached error and nodes don’t get shutdown. >>> >>> The procedure I follow to shut down the nodes are- >>> >>> 1) Unmounts all lustre file systems, >>> >>> 2) Stop lnet service ( lnet service stops successfully.), >>> >>> 3) Unload lustre module, >>> >>> >>> >>> After performing above steps and rebooting node it sucks, error is attached. >>> >>> >>> >>> Regards, >>> >>> Parag >>> >>> >>> >>> >>> >>> >>> ___ >>> lustre-discuss mailing list >>> lustre-discuss@lists.lustre.org >>> http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org >>> >> > >___ >lustre-discuss mailing list >lustre-discuss@lists.lustre.org >http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
Re: [lustre-discuss] lnet shutdown issue while cleint reboot
I cloned from github. Lnetctl is there on lustre servers but on clients only lctl is available. Regards, Parag -Original Message- From: Arman Khalatyan [mailto:arm2...@gmail.com] Sent: Wednesday, January , 2018 7:06 PM To: Parag Khuraswar Cc: Lustre discussion Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot Strange, Is it some custom lustre ? the 2.20 is not yet there: http://lustre.org/download/ lnetctl is inside since 2.10.x On Wed, Jan 3, 2018 at 12:46 PM, Parag Khuraswarwrote: > Hi, > > In my version of lustre on client nodes 'lnetctl " command is not available. > > Regards, > Parag > > > > -Original Message- > From: Arman Khalatyan [mailto:arm2...@gmail.com] > Sent: Wednesday, January , 2018 4:54 PM > To: Parag Khuraswar > Cc: Lustre discussion > Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot > > hi, > Try this before reboot: > umount /lustre > service lnet stop > lnetctl lnet unconfigure > lustre_rmmod > then reboot > On Centos 7.4 it works. > > Cheers, > Arman. > > > On Wed, Jan 3, 2018 at 10:39 AM, Parag Khuraswar > wrote: >> Hi, >> >> >> >> I am using lustre 2.20.1 on RHEL 7.3. On the lustre client nodes when >> I shutdown I get attached error and nodes don’t get shutdown. >> >> The procedure I follow to shut down the nodes are- >> >> 1) Unmounts all lustre file systems, >> >> 2) Stop lnet service ( lnet service stops successfully.), >> >> 3) Unload lustre module, >> >> >> >> After performing above steps and rebooting node it sucks, error is attached. >> >> >> >> Regards, >> >> Parag >> >> >> >> >> >> >> ___ >> lustre-discuss mailing list >> lustre-discuss@lists.lustre.org >> http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org >> > ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
Re: [lustre-discuss] lnet shutdown issue while cleint reboot
Strange, Is it some custom lustre ? the 2.20 is not yet there: http://lustre.org/download/ lnetctl is inside since 2.10.x On Wed, Jan 3, 2018 at 12:46 PM, Parag Khuraswarwrote: > Hi, > > In my version of lustre on client nodes 'lnetctl " command is not available. > > Regards, > Parag > > > > -Original Message- > From: Arman Khalatyan [mailto:arm2...@gmail.com] > Sent: Wednesday, January , 2018 4:54 PM > To: Parag Khuraswar > Cc: Lustre discussion > Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot > > hi, > Try this before reboot: > umount /lustre > service lnet stop > lnetctl lnet unconfigure > lustre_rmmod > then reboot > On Centos 7.4 it works. > > Cheers, > Arman. > > > On Wed, Jan 3, 2018 at 10:39 AM, Parag Khuraswar > wrote: >> Hi, >> >> >> >> I am using lustre 2.20.1 on RHEL 7.3. On the lustre client nodes when >> I shutdown I get attached error and nodes don’t get shutdown. >> >> The procedure I follow to shut down the nodes are- >> >> 1) Unmounts all lustre file systems, >> >> 2) Stop lnet service ( lnet service stops successfully.), >> >> 3) Unload lustre module, >> >> >> >> After performing above steps and rebooting node it sucks, error is attached. >> >> >> >> Regards, >> >> Parag >> >> >> >> >> >> >> ___ >> lustre-discuss mailing list >> lustre-discuss@lists.lustre.org >> http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org >> > ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
Re: [lustre-discuss] Lustre Client in a container
Thanks for all the answers. I was thinking of creating a new file system, starting from clean configuration, implementing quotas etc. For that I was looking for a way in which the systems can coexist, moving symbolic links while the folders are synchronized to the new system. In the process emptying disks of the old file system and moving them to the new one. This is a long process that might take more then a month, but can be done without disturbing normal cluster operation. As it doesn't seems to be possible in real life, I will have to reevaluate my options and come with a different migration schema. On Wed, Jan 3, 2018 at 1:49 PM, Patrick Farrellwrote: > FWIW, as long as you don’t intend to use any interesting features (quotas, > etc), 1.8 clients were used with 2.5 servers at ORNL for some time with no > ill effects on the IO side of things. > > I’m not sure how much further that limited compatibility goes, though. > -- > *From:* Dilger, Andreas > *Sent:* Wednesday, January 3, 2018 4:20:56 AM > *To:* David Cohen > *Cc:* Patrick Farrell; lustre-discuss@lists.lustre.org > *Subject:* Re: [lustre-discuss] Lustre Client in a container > > On Dec 31, 2017, at 01:50, David Cohen > wrote: > > > > Patrick, > > Thanks for you response. > > I looking for a way to migrate from 1.8.9 system to 2.10.2, stable > enough to run the several weeks or more that it might take. > > Note that there is no longer direct support for upgrading from 1.8 to > 2.10. > > That said, are you upgrading the filesystem in place, or are you copying > the data from the 1.8.9 filesystem to the 2.10.2 filesystem? In the latter > case, the upgrade compatibility doesn't really matter. What you need is a > client that can mount both server versions at the same time. > > Unfortunately, no 2.x clients can mount the 1.8.x server filesystem > directly, so that does limit your options. There was a time of > interoperability with 1.8 clients being able to mount 2.1-ish servers, but > that doesn't really help you. You could upgrade the 1.8 servers to 2.1 or > later, and then mount both filesystems with a 2.5-ish client, or upgrade > the servers to 2.5. > > Cheers, Andreas > > > On Sun, Dec 31, 2017 at 12:12 AM, Patrick Farrell wrote: > > David, > > > > I have no direct experience trying this, but I would imagine not - > Lustre is a kernel module (actually a set of kernel modules), so unless the > container tech you're using allows loading multiple different versions of > *kernel modules*, this is likely impossible. My limited understanding of > container tech on Linux suggests that this would be impossible, containers > allow userspace separation but there is only one kernel/set of > modules/drivers. > > > > I don't know of any way to run multiple client versions on the same node. > > > > The other question is *why* do you want to run multiple client versions > on one node...? Clients are usually interoperable across a pretty generous > set of server versions. > > > > - Patrick > > > > > > From: lustre-discuss on > behalf of David Cohen > > Sent: Saturday, December 30, 2017 11:45:15 AM > > To: lustre-discuss@lists.lustre.org > > Subject: [lustre-discuss] Lustre Client in a container > > > > Hi, > > Is it possible to run Lustre client in a container? > > The goal is to run two different client version on the same node, can it > be done? > > > > David > > > > > > ___ > > lustre-discuss mailing list > > lustre-discuss@lists.lustre.org > > http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org > > Cheers, Andreas > -- > Andreas Dilger > Lustre Principal Architect > Intel Corporation > > > > > > > > ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
Re: [lustre-discuss] Lustre Client in a container
FWIW, as long as you don’t intend to use any interesting features (quotas, etc), 1.8 clients were used with 2.5 servers at ORNL for some time with no ill effects on the IO side of things. I’m not sure how much further that limited compatibility goes, though. From: Dilger, AndreasSent: Wednesday, January 3, 2018 4:20:56 AM To: David Cohen Cc: Patrick Farrell; lustre-discuss@lists.lustre.org Subject: Re: [lustre-discuss] Lustre Client in a container On Dec 31, 2017, at 01:50, David Cohen wrote: > > Patrick, > Thanks for you response. > I looking for a way to migrate from 1.8.9 system to 2.10.2, stable enough to > run the several weeks or more that it might take. Note that there is no longer direct support for upgrading from 1.8 to 2.10. That said, are you upgrading the filesystem in place, or are you copying the data from the 1.8.9 filesystem to the 2.10.2 filesystem? In the latter case, the upgrade compatibility doesn't really matter. What you need is a client that can mount both server versions at the same time. Unfortunately, no 2.x clients can mount the 1.8.x server filesystem directly, so that does limit your options. There was a time of interoperability with 1.8 clients being able to mount 2.1-ish servers, but that doesn't really help you. You could upgrade the 1.8 servers to 2.1 or later, and then mount both filesystems with a 2.5-ish client, or upgrade the servers to 2.5. Cheers, Andreas > On Sun, Dec 31, 2017 at 12:12 AM, Patrick Farrell wrote: > David, > > I have no direct experience trying this, but I would imagine not - Lustre is > a kernel module (actually a set of kernel modules), so unless the container > tech you're using allows loading multiple different versions of *kernel > modules*, this is likely impossible. My limited understanding of container > tech on Linux suggests that this would be impossible, containers allow > userspace separation but there is only one kernel/set of modules/drivers. > > I don't know of any way to run multiple client versions on the same node. > > The other question is *why* do you want to run multiple client versions on > one node...? Clients are usually interoperable across a pretty generous set > of server versions. > > - Patrick > > > From: lustre-discuss on behalf of > David Cohen > Sent: Saturday, December 30, 2017 11:45:15 AM > To: lustre-discuss@lists.lustre.org > Subject: [lustre-discuss] Lustre Client in a container > > Hi, > Is it possible to run Lustre client in a container? > The goal is to run two different client version on the same node, can it be > done? > > David > > > ___ > lustre-discuss mailing list > lustre-discuss@lists.lustre.org > http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org Cheers, Andreas -- Andreas Dilger Lustre Principal Architect Intel Corporation ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
Re: [lustre-discuss] lnet shutdown issue while cleint reboot
Hi, In my version of lustre on client nodes 'lnetctl " command is not available. Regards, Parag -Original Message- From: Arman Khalatyan [mailto:arm2...@gmail.com] Sent: Wednesday, January , 2018 4:54 PM To: Parag Khuraswar Cc: Lustre discussion Subject: Re: [lustre-discuss] lnet shutdown issue while cleint reboot hi, Try this before reboot: umount /lustre service lnet stop lnetctl lnet unconfigure lustre_rmmod then reboot On Centos 7.4 it works. Cheers, Arman. On Wed, Jan 3, 2018 at 10:39 AM, Parag Khuraswarwrote: > Hi, > > > > I am using lustre 2.20.1 on RHEL 7.3. On the lustre client nodes when > I shutdown I get attached error and nodes don’t get shutdown. > > The procedure I follow to shut down the nodes are- > > 1) Unmounts all lustre file systems, > > 2) Stop lnet service ( lnet service stops successfully.), > > 3) Unload lustre module, > > > > After performing above steps and rebooting node it sucks, error is attached. > > > > Regards, > > Parag > > > > > > > ___ > lustre-discuss mailing list > lustre-discuss@lists.lustre.org > http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org > ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
Re: [lustre-discuss] lnet shutdown issue while cleint reboot
hi, Try this before reboot: umount /lustre service lnet stop lnetctl lnet unconfigure lustre_rmmod then reboot On Centos 7.4 it works. Cheers, Arman. On Wed, Jan 3, 2018 at 10:39 AM, Parag Khuraswarwrote: > Hi, > > > > I am using lustre 2.20.1 on RHEL 7.3. On the lustre client nodes when I > shutdown I get attached error and nodes don’t get shutdown. > > The procedure I follow to shut down the nodes are- > > 1) Unmounts all lustre file systems, > > 2) Stop lnet service ( lnet service stops successfully.), > > 3) Unload lustre module, > > > > After performing above steps and rebooting node it sucks, error is attached. > > > > Regards, > > Parag > > > > > > > ___ > lustre-discuss mailing list > lustre-discuss@lists.lustre.org > http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org > ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
Re: [lustre-discuss] Lustre Client in a container
On Dec 31, 2017, at 01:50, David Cohenwrote: > > Patrick, > Thanks for you response. > I looking for a way to migrate from 1.8.9 system to 2.10.2, stable enough to > run the several weeks or more that it might take. Note that there is no longer direct support for upgrading from 1.8 to 2.10. That said, are you upgrading the filesystem in place, or are you copying the data from the 1.8.9 filesystem to the 2.10.2 filesystem? In the latter case, the upgrade compatibility doesn't really matter. What you need is a client that can mount both server versions at the same time. Unfortunately, no 2.x clients can mount the 1.8.x server filesystem directly, so that does limit your options. There was a time of interoperability with 1.8 clients being able to mount 2.1-ish servers, but that doesn't really help you. You could upgrade the 1.8 servers to 2.1 or later, and then mount both filesystems with a 2.5-ish client, or upgrade the servers to 2.5. Cheers, Andreas > On Sun, Dec 31, 2017 at 12:12 AM, Patrick Farrell wrote: > David, > > I have no direct experience trying this, but I would imagine not - Lustre is > a kernel module (actually a set of kernel modules), so unless the container > tech you're using allows loading multiple different versions of *kernel > modules*, this is likely impossible. My limited understanding of container > tech on Linux suggests that this would be impossible, containers allow > userspace separation but there is only one kernel/set of modules/drivers. > > I don't know of any way to run multiple client versions on the same node. > > The other question is *why* do you want to run multiple client versions on > one node...? Clients are usually interoperable across a pretty generous set > of server versions. > > - Patrick > > > From: lustre-discuss on behalf of > David Cohen > Sent: Saturday, December 30, 2017 11:45:15 AM > To: lustre-discuss@lists.lustre.org > Subject: [lustre-discuss] Lustre Client in a container > > Hi, > Is it possible to run Lustre client in a container? > The goal is to run two different client version on the same node, can it be > done? > > David > > > ___ > lustre-discuss mailing list > lustre-discuss@lists.lustre.org > http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org Cheers, Andreas -- Andreas Dilger Lustre Principal Architect Intel Corporation ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
[lustre-discuss] lnet shutdown issue while cleint reboot
Hi, I am using lustre 2.20.1 on RHEL 7.3. On the lustre client nodes when I shutdown I get attached error and nodes don't get shutdown. The procedure I follow to shut down the nodes are- 1) Unmounts all lustre file systems, 2) Stop lnet service ( lnet service stops successfully.), 3) Unload lustre module, After performing above steps and rebooting node it sucks, error is attached. Regards, Parag ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org