August 2019 - ceph-users

Migrate RBD header and metadata for EC backed image to new pool
by Jason Lorsung 05 Aug '19

05 Aug '19

I'm looking at decommissioning a pool (rbd) that is the header and metadata storage pool for a number of ec backed images (created with --data-pool ecrbd). Pool ecrbd is remaining intact so I don't want to have to migrate all the image data if unnecessary. Is there a way to migrate all the header and metadata off of pool rbd to pool rbdnew, while leaving all the data inplace?

1 0

OSD crashes on start - while loading pgs
by Jesper Stemann Andersen 03 Aug '19

03 Aug '19

Hi, Any ideas for resolving an issue where an OSD crashes on start-up? I have one (large hdd) OSD that will no longer start – it crashes while loading pgs - see attached log file - excerpt below: 2019-08-02 10:08:21.021207 7fea86d7be00 0 osd.1 1844 load_pgs 2019-08-02 10:08:39.370112 7fea86d7be00 -1 *** Caught signal (Aborted) ** in thread 7fea86d7be00 thread_name:ceph-osd ceph version 12.2.12 (39cfebf25a7011204a9876d2950e4b28aba66d11) luminous (stable) 1: (()+0xa59c94) [0x55b835a6dc94] 2: (()+0x110e0) [0x7fea843800e0] 3: (gsignal()+0xcf) [0x7fea83347fff] 4: (abort()+0x16a) [0x7fea8334942a] 5: (__gnu_cxx::__verbose_terminate_handler()+0x15d) [0x7fea83c600ad] 6: (()+0x8f066) [0x7fea83c5e066] 7: (()+0x8f0b1) [0x7fea83c5e0b1] 8: (()+0x8f2c9) [0x7fea83c5e2c9] 9: (pg_log_entry_t::decode_with_checksum(ceph::buffer::list::iterator&)+0x156) [0x55b8356f57c6] 10: (void PGLog::read_log_and_missing<pg_missing_set<true> >(ObjectStore*, coll_t, coll_t, ghobject_t, pg_info_t const&, PGLog::IndexedLog&, pg_missing_set<true>&, bool, std::__cxx11::basic_ostringstream<char, std::char_traits<char>, std::allocator<char> >&, bool, bool*, DoutPrefixProvider const*, std::set<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > >*, bool)+0x1ab4) [0x55b8355a6584] 11: (PG::read_state(ObjectStore*, ceph::buffer::list&)+0x38b) [0x55b83554b7eb] 12: (OSD::load_pgs()+0x8b8) [0x55b835496678] 13: (OSD::init()+0x2237) [0x55b8354b75c7] 14: (main()+0x3092) [0x55b8353bf1c2] 15: (__libc_start_main()+0xf1) [0x7fea833352e1] 16: (_start()+0x2a) [0x55b83544b8ca] I have ensured that kernel.pid_max is set to a high value – sysctl reports kernel.pid_max = 4194304 This issue arose following an expansion of the ceph cluster: https://forum.proxmox.com/threads/unable-to-start-osd-crashes-while-loading… In summary: I added a third node, with extra OSD’s, and increased pg_num and pgp_num for one pool before the cluster had settled. However, by now the cluster has settled – I no longer have the global setting mon_max_pg_per_osd = 1000. Only the issue with the OSD that will not start remains. Best regards, Jesper Stemann Andersen Lead Software Engineer, R&D IHP Systems A/S +45 26 25 23 91

1 0

Help needed pls (rbd cache)
by Muhammad Junaid 02 Aug '19

02 Aug '19

Hi there, Sorry for asking a question, which may be of very basic nature and asked many times before. But much Google search can not satisfy me. The question is about RBD Cache in write-back mode using KVM/libvirt. If we enable this, it uses local KVM Host's RAM as cache for VM's write requests. And KVM Host immediately responds to VM's OS that data has been written to Disk (Actually it is still not on OSD's yet). Then how can be it power failure safe? Is my understanding correct? If not, pls correct. This is very important for me. Thank you very much in advance. Best regards. Muhammad Junaid

2 2

Ceph packages for Debian buster
by yuval.freund＠cloud.ionos.com 01 Aug '19

01 Aug '19

We've been using ceph on Debian for several years now, and would really like seeing official packages for buster. We recently resorted to building our own Luminous packages for buster (based on the Ubuntu packages), due to the fact that: * We've already upgraded our clusters to 12.2.12 , and the debian.org packages offer 12.2.11 . * The official ceph packages for stretch are built against libcurl3, which has been removed in buster in favor of libcurl4 . Sage stated last year that Luminous and Nautilus (and possibly Mimic) packages for buster should be provided at some point (http://lists.ceph.com/pipermail/ceph-users-ceph.com/2018-June/027478.html) and we really hope that this decision hasn't been reverted - a lot of Debian users are looking forward to being able to upgrade their clusters to a newer release. We also believe having official Luminous packages will allow most Debian users who've kept their systems up-to-date to have a clean upgrade path. We are aware of croit's Nautilus packages for buster and have played around with them (thanks!), but would rather go with official packages in our production clusters, preferably built against python3 (for bonus points ;) ). We're also undecided regarding whether we should go straight to Nautilus, mainly due to the fact Mimic seems like the more stable choice for the time being. We'd be happy and willing to assist in any way we can. Cheers, Yuval -- Yuval Freund System Engineer 1&1 IONOS Cloud GmbH | Greifswalder Str. 207 | 10405 Berlin | Germany Web: www.ionos.de Head Office: Berlin, Germany District Court Berlin Charlottenburg, Registration number: HRB 125506 B Executive Management: Christoph Steffens, Matthias Steinberg, Achim Weiss Member of United Internet

1 0