when an osd is started up, IO will be blocked
Hi all, When an osd is started, relative IO will be blocked. According to the test result,the larger iops the clients send , the longer it will take to elapse. Adjustment on all the parameters associate with recovery operations was also found useless. How to reduce the impact of this process on the IO ? Thanks and Regards, WangSongbo -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: when an osd is started up, IO will be blocked
, 5.00th=[4], 10.00th=[5], 20.00th=[6], | 30.00th=[6], 40.00th=[7], 50.00th=[8], 60.00th=[9], | 70.00th=[ 10], 80.00th=[ 12], 90.00th=[ 14], 95.00th=[ 709], | 99.00th=[16712], 99.50th=[16712], 99.90th=[16712], 99.95th=[16712], | 99.99th=[16712] bw (KB /s): min= 129, max= 2038, per=100.00%, avg=774.00, stdev=1094.73 write: io=24976KB, bw=401264B/s,*iops=48,* runt= 63737msec slat (usec): min=0, max=40, avg= 2.48, stdev= 3.30 clat (msec): min=2, max=31379, avg=786.91, stdev=4829.02 lat (msec): min=2, max=31379, avg=786.92, stdev=4829.02 clat percentiles (msec): | 1.00th=[4], 5.00th=[6], 10.00th=[6], 20.00th=[8], | 30.00th=[9], 40.00th=[9], 50.00th=[ 11], 60.00th=[ 12], | 70.00th=[ 13], 80.00th=[ 15], 90.00th=[ 19], 95.00th=[ 29], | 99.00th=[16712], 99.50th=[16712], 99.90th=[16712], 99.95th=[16712], | 99.99th=[16712] bw (KB /s): min= 317, max= 5228, per=100.00%, avg=1957.33, stdev=2832.48 lat (msec) : 2=0.07%, 4=3.23%, 10=52.29%, 20=35.76%, 50=4.41% lat (msec) : 750=1.36%, 1000=0.02%, 2000=0.02%, >=2000=2.83% cpu : usr=0.03%, sys=0.00%, ctx=228, majf=0, minf=19 IO depths: 1=0.1%, 2=0.1%, 4=0.4%, 8=3.9%, 16=18.9%, 32=73.1%, >=64=3.5% submit: 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0% complete : 0=0.0%, 4=97.5%, 8=0.0%, 16=0.2%, 32=0.9%, 64=1.5%, >=64=0.0% issued: total=r=1363/w=3122/d=0, short=r=0/w=0/d=0, drop=r=0/w=0/d=0 latency : target=0, window=0, percentile=100.00%, depth=64 Run status group 0 (all jobs): READ: io=10904KB, aggrb=171KB/s, minb=171KB/s, maxb=171KB/s, mint=63737msec, maxt=63737msec WRITE: io=24976KB, aggrb=391KB/s, minb=391KB/s, maxb=391KB/s, mint=63737msec, maxt=63737msec Disk stats (read/write): dm-0: ios=81/34, merge=0/0, ticks=472/26, in_queue=498, util=0.30%, aggrios=143/102, aggrmerge=106/122, aggrticks=1209/134, aggrin_queue=1343, aggrutil=0.60% sdd: ios=143/102, merge=106/122, ticks=1209/134, in_queue=1343, util=0.60% Thanks and Regards, WangSongbo On 15/10/22 下午10:43, wangsongbo wrote: Hi all, When an osd is started, relative IO will be blocked. According to the test result,the larger iops the clients send , the longer it will take to elapse. Adjustment on all the parameters associate with recovery operations was also found useless. How to reduce the impact of this process on the IO ? Thanks and Regards, WangSongbo -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
How to reduce the influenct on the IO when an osd is marked out?
Hi all, when an osd is marked out, relative IO will be blocked, in which case, application built on ceph will fail.According to test result, the larger a data is,the longer it will take to elapse. How to reduce the impact of this process on the IO? Thanks and Regards, WangSongbo -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
failed to open http://apt-mirror.front.sepia.ceph.com
Hi Loic and other Cephers, I am running teuthology-suites in our testing, because the connection to "apt-mirror.front.sepia.ceph.com" timed out , "ceph-cm-ansible" failed. And from a web-browser, I got the response like this : "502 Bad Gateway". "64.90.32.37 apt-mirror.front.sepia.ceph.com" has been added to /etc/hosts. Did the resources have been removed ? Thanks and Regards, WangSongbo -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: failed to open http://apt-mirror.front.sepia.ceph.com
Sage and Loic, Thanks for your reply. I am running teuthology in our testing.I can send a traceroute to 64.90.32.37. but when ceph-cm-ansible run the " yum-complete-transaction --cleanup-only" command, it got such a response :"http://apt-mirror.front.sepia.ceph.com/misc-rpms/repodata/repomd.xml: [Errno 14] PYCURL ERROR 7 - "Failed connect to apt-mirror.front.sepia.ceph.com:80; Connection timed out" And I replace "apt-mirror.front.sepia.ceph.com" to "64.90.32.37" in repo file, then run "yum-complete-transaction --cleanup-only" command, I got a response like this:"http://64.90.32.37/misc-rpms/repodata/repomd.xml: [Errno 14] PYCURL ERROR 22 - "The requested URL returned error: 502 Bad Gateway"" I do not know whether it was affected by the last week's attack. Thanks and Regards, WangSongbo On 15/9/23 下午11:22, Loic Dachary wrote: On 23/09/2015 15:11, Sage Weil wrote: On Wed, 23 Sep 2015, Loic Dachary wrote: Hi, On 23/09/2015 12:29, wangsongbo wrote: 64.90.32.37 apt-mirror.front.sepia.ceph.com It works for me. Could you send a traceroute apt-mirror.front.sepia.ceph.com ? This is a private IP internal to the sepia lab. Anythign outside the lab shouldn't be using it... This is the public facing IP and is required for teuthology to run outside of the lab (http://tracker.ceph.com/issues/12212). 64.90.32.37 apt-mirror.front.sepia.ceph.com suggests the workaround was used. And a traceroute will confirm if the resolution happens as expected (with the public IP) or with a private IP (meaning the workaround is not in place where it should). Cheers -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [ceph-users] Important security noticed regarding release signing key
Hi Ken, Thanks for your reply. But in the ceph-cm-ansible project scheduled by teuthology, "ceph.com/packages/ceph-extras" is in used now, such as qemu-kvm-0.12.1.2-2.415.el6.3ceph, qemu-kvm-tools-0.12.1.2-2.415.el6.3ceph etc. Any new releases will be provided ? On 15/9/22 下午10:24, Ken Dreyer wrote: On Tue, Sep 22, 2015 at 2:38 AM, Songbo Wang <songbo1...@gmail.com> wrote: Hi, all, Since the last week‘s attack, “ceph.com/packages/ceph-extras” can be opened never, but where can I get the releases of ceph-extra now? Thanks and Regards, WangSongbo The packages in "ceph-extras" were old and subject to CVEs (the big one being VENOM, CVE-2015-3456). So I don't intend to host ceph-extras in the new location. - Ken -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [ceph-users] Important security noticed regarding release signing key
Hi Ken, Thank you, I will update my repo and continue my test. - Songbo On 15/9/23 上午10:50, Ken Dreyer wrote: Hi Songbo, It's been removed from Ansible now: https://github.com/ceph/ceph-cm-ansible/pull/137 - Ken On Tue, Sep 22, 2015 at 8:33 PM, wangsongbo <songbo1...@gmail.com> wrote: Hi Ken, Thanks for your reply. But in the ceph-cm-ansible project scheduled by teuthology, "ceph.com/packages/ceph-extras" is in used now, such as qemu-kvm-0.12.1.2-2.415.el6.3ceph, qemu-kvm-tools-0.12.1.2-2.415.el6.3ceph etc. Any new releases will be provided ? On 15/9/22 下午10:24, Ken Dreyer wrote: On Tue, Sep 22, 2015 at 2:38 AM, Songbo Wang <songbo1...@gmail.com> wrote: Hi, all, Since the last week‘s attack, “ceph.com/packages/ceph-extras” can be opened never, but where can I get the releases of ceph-extra now? Thanks and Regards, WangSongbo The packages in "ceph-extras" were old and subject to CVEs (the big one being VENOM, CVE-2015-3456). So I don't intend to host ceph-extras in the new location. - Ken -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
test
test -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
the release of ceph-extra
Hi, all, Since the last week‘s attack, “ceph.com/packages/ceph-extras <http://ceph.com/packages/ceph-extras>” can be opened never, but where can I get the releases of ceph-extra now? Thanks and Regards, WangSongbo -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: failed to open http://apt-mirror.front.sepia.ceph.com
Loic, It's my fault. The dns server I set is unreachable. when I modify that , everything is ok. Thanks and Regards, WangSongbo On 15/9/24 上午1:01, Loic Dachary wrote: On 23/09/2015 18:50, wangsongbo wrote: Sage and Loic, Thanks for your reply. I am running teuthology in our testing.I can send a traceroute to 64.90.32.37. but when ceph-cm-ansible run the " yum-complete-transaction --cleanup-only" command, it got such a response :"http://apt-mirror.front.sepia.ceph.com/misc-rpms/repodata/repomd.xml: [Errno 14] PYCURL ERROR 7 - "Failed connect to apt-mirror.front.sepia.ceph.com:80; Connection timed out" And I replace "apt-mirror.front.sepia.ceph.com" to "64.90.32.37" in repo file, then run "yum-complete-transaction --cleanup-only" command, I got a response like this:"http://64.90.32.37/misc-rpms/repodata/repomd.xml: [Errno 14] PYCURL ERROR 22 - "The requested URL returned error: 502 Bad Gateway"" I do not know whether it was affected by the last week's attack. Querying the IP directly won't get you where the mirror is (it's a vhost). I think ansible fails because it queries the DNS and does not use the entry you set in the /etc/hosts file. The OpenStack teuthology backend sets a specific entry in the DNS to workaround the problem (see https://github.com/ceph/teuthology/blob/master/teuthology/openstack/setup-openstack.sh#L318) Cheers Thanks and Regards, WangSongbo On 15/9/23 下午11:22, Loic Dachary wrote: On 23/09/2015 15:11, Sage Weil wrote: On Wed, 23 Sep 2015, Loic Dachary wrote: Hi, On 23/09/2015 12:29, wangsongbo wrote: 64.90.32.37 apt-mirror.front.sepia.ceph.com It works for me. Could you send a traceroute apt-mirror.front.sepia.ceph.com ? This is a private IP internal to the sepia lab. Anythign outside the lab shouldn't be using it... This is the public facing IP and is required for teuthology to run outside of the lab (http://tracker.ceph.com/issues/12212). 64.90.32.37 apt-mirror.front.sepia.ceph.com suggests the workaround was used. And a traceroute will confirm if the resolution happens as expected (with the public IP) or with a private IP (meaning the workaround is not in place where it should). Cheers -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [ceph-users] Important security noticed regarding release signing key
Hi Ken, Just now, I run teuthology-suites in our testing, it failed because of lacking these packages, such as qemu-kvm-0.12.1.2-2.415.el6.3ceph.x86_64, qemu-kvm-tools-0.12.1.2-2.415.el6.3ceph etc. The modify "rm ceph-extras repository config#137" only remove the repository , but did not solve the ansible's dependence. How to solve this dependence ? Thanks and Regards, WangSongbo On 15/9/23 上午10:50, Ken Dreyer wrote: Hi Songbo, It's been removed from Ansible now: https://github.com/ceph/ceph-cm-ansible/pull/137 - Ken On Tue, Sep 22, 2015 at 8:33 PM, wangsongbo <songbo1...@gmail.com> wrote: Hi Ken, Thanks for your reply. But in the ceph-cm-ansible project scheduled by teuthology, "ceph.com/packages/ceph-extras" is in used now, such as qemu-kvm-0.12.1.2-2.415.el6.3ceph, qemu-kvm-tools-0.12.1.2-2.415.el6.3ceph etc. Any new releases will be provided ? On 15/9/22 下午10:24, Ken Dreyer wrote: On Tue, Sep 22, 2015 at 2:38 AM, Songbo Wang <songbo1...@gmail.com> wrote: Hi, all, Since the last week‘s attack, “ceph.com/packages/ceph-extras” can be opened never, but where can I get the releases of ceph-extra now? Thanks and Regards, WangSongbo The packages in "ceph-extras" were old and subject to CVEs (the big one being VENOM, CVE-2015-3456). So I don't intend to host ceph-extras in the new location. - Ken -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
The osd process locked itself , when I tested cephfs through filebench
inLockDelay(int volatile*, int, int) () from /usr/lib64/libtcmalloc.so.4 #1 0x7fa9b772f9bc in SpinLock::SlowLock() () from /usr/lib64/libtcmalloc.so.4 #2 0x7fa9b772c0f1 in ?? () from /usr/lib64/libtcmalloc.so.4 #3 0x7fa9b7725626 in MallocHook::InvokeNewHookSlow(void const*, unsigned long) () from /usr/lib64/libtcmalloc.so.4 #4 0x7fa9b7732d73 in tc_new () from /usr/lib64/libtcmalloc.so.4 #5 0x00b2b982 in ceph::log::Log::create_entry (this=optimized out>, level=0, subsys=27) at log/Log.cc:175 #6 0x00c03075 in Pipe::fault (this=0x3b53d800, onread=optimized out>) at msg/simple/Pipe.cc:1392 #7 0x00c12114 in Pipe::reader (this=0x3b53d800) at msg/simple/Pipe.cc:1674 #8 0x00c1606d in Pipe::Reader::entry (this=out>) at msg/simple/Pipe.h:50 #9 0x7fa9b6f69a51 in start_thread () from /lib64/libpthread.so.0 #10 0x7fa9b5cf193d in clone () from /lib64/libc.so.6 (gdb) <- This is not the first time for me to find above problem. Thanks, wangsongbo -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html