Re: [sheepdog] Is sheepdog still actively maintained?

It not maintained. пт, 15 мар. 2019 г., 15:10 Christian Ebner : > Hello, > > I was wondering if the sheepdog project is still actively maintained? > According to the commit frequency on the github repo, development is > strongly decreased over the last few months. > > Any feedback from the

Re: [sheepdog] [Qemu-devel] internal snapshots with sheepdog

but when i'm try in libvirt create live snapshot - libvirt says, that this is not possible with raw image. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] Live snapshot not working via libvirt. Possible regression?

ry. Hitoshi, when the sheepdog announce this feature hot qemu store memory inside sheepdog ? Does qemu create additional volume for memory or something else ? -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] Live snapshot not working via libvirt. Possible regression?

2015-09-11 10:43 GMT+03:00 Vasiliy Tolstov <v.tols...@selfip.ru>: > > As i'm understand commit messae, when we create vdi with 20 Gb size > and want to to create full snapshot (with disk and memory), qemu needs > additional space to store memory. But all disk size from vdi a

Re: [sheepdog] [PATCH 2/2] sheepdog: refine discard support

04 сент. 2015 г. 12:57 пользователь "Hitoshi Mitake" < mitake.hito...@gmail.com> написал: > > On Fri, Sep 4, 2015 at 5:51 PM, Hitoshi Mitake <mitake.hito...@gmail.com> wrote: > > On Wed, Sep 2, 2015 at 9:36 PM, Vasiliy Tolstov <v.tols...@selfip.ru> wrote

Re: [sheepdog] [PATCH 2/2] sheepdog: refine discard support

ki.teru...@lab.ntt.co.jp> > Cc: Vasiliy Tolstov <v.tols...@selfip.ru> > Cc: Jeff Cody <jc...@redhat.com> > Signed-off-by: Hitoshi Mitake <mitake.hito...@lab.ntt.co.jp> I'm test this patch and now discard working properly and no errors in sheepdog log file. Tested-b

Re: [sheepdog] [PATCH 1/2] sheepdog: use per AIOCB dirty indexes for non overlapping requests

h also does trivial renaming for better description: > overwrapping -> overlapping > > Cc: Teruaki Ishizaki <ishizaki.teru...@lab.ntt.co.jp> > Cc: Vasiliy Tolstov <v.tols...@selfip.ru> > Cc: Jeff Cody <jc...@redhat.com> > Signed-off-by: Hitoshi Mitake <mi

Re: [sheepdog] [sheepdog/sheepdog] c4e320: shared lib: drop unnecessary return value check

request status. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] [Qemu-devel] [PATCH] sheepdog: fix overlapping metadata update

with debug symbols? qemu-system-x86[34046]: segfault at 401364 ip 7f33f52a1ff8 sp 7f3401ecad30 error 4 in qemu-system-x86_64[7f33f4efd000+518000] -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] [Qemu-devel] [PATCH] sheepdog: fix overlapping metadata update

how to determine qemu segfault in my case, does it possible to build qemu with debugging via configure flags ang get more detailed descriptions? Thanks, Yuan -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman

Re: [sheepdog] [Qemu-devel] [PATCH] sheepdog: fix overlapping metadata update

2015-07-31 15:08 GMT+03:00 Vasiliy Tolstov v.tols...@selfip.ru: Please wait to performance comparison. As i see Liu's patch may be more slow then Hitoshi. I'm switch to local cluster driver to test only local ssd and not network overhead. But now Liu's qemu does not able to completely install

Re: [sheepdog] [Qemu-devel] [PATCH] sheepdog: fix overlapping metadata update

, aggrticks=1072368/978452, aggrin_queue=2052424, aggrutil=35.39% sdd: ios=184963/184383, merge=0/3, ticks=1072368/978452, in_queue=2052424, util=35.39% 2015-08-02 14:52 GMT+03:00 Vasiliy Tolstov v.tols...@selfip.ru: 2015-07-31 15:08 GMT+03:00 Vasiliy Tolstov v.tols...@selfip.ru: Please wait

[sheepdog] [sheepdog/sheepdog] 3e1292: fixe incorrect man and help for vdi backup/restore

Branch: refs/heads/master Home: https://github.com/sheepdog/sheepdog Commit: 3e129212d9ea0e60eba3baa775580ee23baade68 https://github.com/sheepdog/sheepdog/commit/3e129212d9ea0e60eba3baa775580ee23baade68 Author: Vasiliy Tolstov v.tols...@selfip.ru Date: 2015-08-02 (Sun, 02 Aug

Re: [sheepdog] [Qemu-devel] [PATCH] sheepdog: fix overlapping metadata update

patch breaks when using discards in qemu =(. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] [Qemu-devel] [PATCH] sheepdog: fix overlapping metadata update

2015-07-31 14:55 GMT+03:00 Vasiliy Tolstov v.tols...@selfip.ru: Liu's patch also works for me. But also like in Hitoshi patch breaks when using discards in qemu =(. Please wait to performance comparison. As i see Liu's patch may be more slow then Hitoshi. -- Vasiliy Tolstov, e-mail: v.tols

Re: [sheepdog] [Qemu-devel] [PATCH] sheepdog: fix overlapping metadata update

. If you have problem with applying the patch, feel free to mail me and I'll package a patched QEMU tree for you. Ok, i'm check you patch too. Please wait. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo

Re: [sheepdog] [Qemu-devel] [PATCH] sheepdog: fix overlapping metadata update

system and corrupt it. After the patch i don't see issues. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] [Qemu-devel] [PATCH] sheepdog: serialize requests to overwrapping area

2015-07-27 18:23 GMT+03:00 Jeff Cody jc...@redhat.com: Thanks, applied to my block branch: https://github.com/codyprime/qemu-kvm-jtc/tree/block Thanks! Waiting for adding to qemu rc =) -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https

Re: [sheepdog] RFC: migrating development process from mailing list to github pull request

Very good news! Tracking patches in pull requests and bugs in issues very easy and informative. Does it possible also cleanup old issues from github? And maybe subscribe this list to sheepdog repo? 24 июля 2015 г. 7:59 пользователь Hitoshi Mitake mitake.hito...@gmail.com написал: Hi sheepdog

Re: [sheepdog] multiply address problem

24 июля 2015 г. 7:29 пользователь Hitoshi Mitake mitake.hito...@gmail.com написал: At Mon, 20 Jul 2015 19:01:12 +0300, Vasiliy Tolstov wrote: [1 text/plain; UTF-8 (quoted-printable)] Also, why not libfabric? I don't know about libfabric, so I don't have any opinions about

Re: [sheepdog] RFC: migrating development process from mailing list to github pull request

or github issues? Launchpad more ubuntu centric and i don't know how to subscribe to all current bugs and feature. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] multiply address problem

ip stack) BTW, why do you need RDMA in the near future? Isn't it possible to use other fast networks (e.g. 10GbE)? This is not big problem now (i'm switch temporarily to bonding with two ib ports). -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog

Re: [sheepdog] multiply address problem

for a while for initial support? Then we will be able to have a discussion about the problem of multiple addresses. Sorry for my late announce... Also what estimated date for this? Because i'm very need this feature for my current work and can't wait 1-2 mounts =(. -- Vasiliy Tolstov, e-mail

[sheepdog] sheepdog + libvirt problems

21494125047808 0% 343597383680 real0m0.298s user0m0.000s sys 0m0.012s -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

[sheepdog] multiply address problem

. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] multiply address problem

for a while for initial support? Then we will be able to have a discussion about the problem of multiple addresses. Sorry for my late announce... Thanks, Hitoshi No problem =) I'm study rdma and learn it. But where i can find such patch? -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru

Re: [sheepdog] multiply address problem

Also, why not libfabric? 20 июля 2015 г. 12:07 пользователь Vasiliy Tolstov v.tols...@selfip.ru написал: 2015-07-20 12:02 GMT+03:00 Hitoshi Mitake mitake.hito...@gmail.com: I'm writing a patch for supporting RDMA with the library from mellanox named accelio: https://github.com/accelio

Re: [sheepdog] filesystem errors when using sheepdog

the branch? Yes, thanks. I'm already waiting for qemu rebuild and test this patches. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] filesystem errors when using sheepdog

2015-07-17 14:55 GMT+03:00 Vasiliy Tolstov v.tols...@selfip.ru: Yes, thanks. I'm already waiting for qemu rebuild and test this patches. with this patches i have qemu segfaults, i can't check in gdb now, but before this i don't have such segfaults: Fri Jul 17 15:20:51 2015] qemu-system-x86

Re: [sheepdog] filesystem errors when using sheepdog

2015-07-17 15:34 GMT+03:00 Hitoshi Mitake mitake.hito...@gmail.com: Oops, sorry, I found a trivial mistake. Updated serialize-v4 branch, could you try again? I don't see second patch in this branch, also update time on it 7 hours ago, does this right? If so - i'm rebuild -- Vasiliy Tolstov

Re: [sheepdog] filesystem errors when using sheepdog

option discard=ignore, log does not have any errors, so i think all works fine. But if that possible - best of all - fix discards =), because it greatly reduce disk usage. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org

Re: [sheepdog] filesystem errors when using sheepdog

patch without enabling discard? yes it solved, if you send to qemu-devel may be this patch goes to 2.4.0. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] filesystem errors when using sheepdog

cfd2d6001e, No object found Jul 17 16:24:00 ERROR [io 7195] sd_remove_object(556) failed to remove object cfd2d6001e, No object found Jul 17 16:24:00 ERROR [io 7195] update_obj_refcnt(600) fail, 2 -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https

Re: [sheepdog] filesystem errors when using sheepdog

not helps i'm tr to rewrite vm installer and share it in boxed variant and send instructions how to run it in qemu. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] filesystem errors when using sheepdog

error log, so I think the problem can happen without discard in the previous qemu. https://gist.githubusercontent.com/vtolstov/c15a1c3c6f2d2d6e72c4/raw My sheep.log that i send does not contains discard, because vm started without discards. So this patch fixes reported issue. -- Vasiliy Tolstov, e

Re: [sheepdog] filesystem errors when using sheepdog

2015-07-17 19:29 GMT+03:00 Hitoshi Mitake mitake.hito...@gmail.com: Thanks, I'll post a patch to qemu list. Can I add your Tested-by tag? Yes, -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] filesystem errors when using sheepdog

2015-07-17 19:47 GMT+03:00 Hitoshi Mitake mitake.hito...@gmail.com: Sorry, I sent the patch without your tested-by tag. Could you reply to the patch with your tested-by tag? No problem, i'm reply to patch email with my tested-by -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog

Re: [sheepdog] filesystem errors when using sheepdog

Can I try patch?which branch? 16 июля 2015 г. 17:34 пользователь Hitoshi Mitake mitake.hito...@gmail.com написал: At Thu, 16 Jul 2015 00:54:47 +0300, Vasiliy Tolstov wrote: 2015-07-15 19:24 GMT+03:00 Hitoshi Mitake mitake.hito...@gmail.com: Thanks a lot for your testing. How about

Re: [sheepdog] filesystem errors when using sheepdog

And thanks for help 16 июля 2015 г. 17:34 пользователь Hitoshi Mitake mitake.hito...@gmail.com написал: At Thu, 16 Jul 2015 00:54:47 +0300, Vasiliy Tolstov wrote: 2015-07-15 19:24 GMT+03:00 Hitoshi Mitake mitake.hito...@gmail.com: Thanks a lot for your testing. How about this branch

Re: [sheepdog] filesystem errors when using sheepdog

,if=none,id=drive-scsi0-0-0-0,format=raw,cache=none,discard=ignore,aio=native -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] filesystem errors when using sheepdog

buggy. I can't write anything on vdi. sheep does not have errors: but full log is there https://gist.githubusercontent.com/vtolstov/f4d34174c2294aee6115/raw qemu log is empty i'm try preallocate vdi but nothing changed -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list

Re: [sheepdog] filesystem errors when using sheepdog

2015-07-15 15:49 GMT+03:00 Vasiliy Tolstov v.tols...@selfip.ru: Thanks, but error not fixed complete sheep.log with debug https://gist.github.com/raw/c15a1c3c6f2d2d6e72c4 -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org

Re: [sheepdog] filesystem errors when using sheepdog

. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] filesystem errors when using sheepdog

. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] filesystem errors when using sheepdog

2015-07-06 16:05 GMT+03:00 Vasiliy Tolstov v.tols...@selfip.ru: Thanks! In my case this is very critical problem, how to fix it? Any news? -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] filesystem errors when using sheepdog

2015-07-04 0:29 GMT+03:00 Vasiliy Tolstov v.tols...@selfip.ru: qemu 2.3.0, also i have this errors with io=threads and cache=writeback I'm debug this isse and qemu message correct: wheepdog returns SD_RES_NO_OBJ when try to write object and qemu displays No object found to stderr, but why

Re: [sheepdog] filesystem errors when using sheepdog

problem but I'm curious). I have 6 disk on each node and want to use multidisk sheepdog feature, so for metadata i create small lvm on each disk and add it to raid1, so sheepdog saves meatadata on raid1, all other space formatted to ext4. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog

Re: [sheepdog] filesystem errors when using sheepdog

2015-07-06 15:53 GMT+03:00 Hitoshi Mitake mitake.hito...@gmail.com: Thanks for your report. Your problem seems to be similar to the problem found in our test: https://bugs.launchpad.net/sheepdog-project/+bug/1456421 Thanks! In my case this is very critical problem, how to fix it? -- Vasiliy

Re: [sheepdog] download vdi via object storage

in files under sheep/http/. Thanks! -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

[sheepdog] alibaba sheepdog question

if a hard disk failed. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

[sheepdog] [PATCH v3] add libcorosync_common depen

when build with corosync 2.x libcorosync_common needs to be linked Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru --- configure.ac | 10 +++--- sheep/Makefile.am | 6 +++--- 2 files changed, 10 insertions(+), 6 deletions(-) diff --git a/configure.ac b/configure.ac index adc3dc1

Re: [sheepdog] [PATCH v3] add libcorosync_common depen

2015-07-03 17:37 GMT+03:00 Vasiliy Tolstov v.tols...@selfip.ru: when build with corosync 2.x libcorosync_common needs to be linked This version does not have bug with [] simbols in PKG_CHECK_MODULES -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog

Re: [sheepdog] filesystem errors when using sheepdog

2015-07-04 0:11 GMT+03:00 Vasiliy Tolstov v.tols...@selfip.ru: I'm using latest sheepdog fomr master branch. Qemu runs with io=native and cache=none sheepdog cluster formatted with -c 2 and store format tree. Sometimes when i'm doing write (5Gb data inside virtual machine) i have file system

[sheepdog] [PATCH v2] add libcorosync_common depen

when build with corosync 2.x libcorosync_common needs to be linked Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru --- configure.ac | 10 +++--- sheep/Makefile.am | 6 +++--- 2 files changed, 10 insertions(+), 6 deletions(-) diff --git a/configure.ac b/configure.ac index adc3dc1

Re: [sheepdog] [[PATCH v1] ] add libcorosync_common depen

. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] download vdi via object storage

2015-07-02 14:52 GMT+03:00 Vasiliy Tolstov v.tols...@selfip.ru: OK, can you share some info about how can I get all vdi pieces, and what files I need to modify? Also offtopic, what cache modes in qemu safe to use in sheepdog? ping... -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru

Re: [sheepdog] download vdi via object storage

02 июля 2015 г. 5:32 пользователь Hitoshi Mitake mitake.hito...@lab.ntt.co.jp написал: At Wed, 1 Jul 2015 17:54:09 +0300, Vasiliy Tolstov wrote: Does it possible to create vdi for vm and after some time download it via swift like sheepdog object storage? How can i navigate link for vdi

Re: [sheepdog] download vdi via object storage

OK, can you share some info about how can I get all vdi pieces, and what files I need to modify? Also offtopic, what cache modes in qemu safe to use in sheepdog? 02 июля 2015 г. 14:49 пользователь Hitoshi Mitake mitake.hito...@gmail.com написал: On Thu, Jul 2, 2015 at 7:37 PM, Vasiliy Tolstov

[sheepdog] [[PATCH v1] ] add libcorosync_common depen

when build with corosync 2.x libcorosync_common needs to be linked Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru --- configure.ac | 1 + sheep/Makefile.am | 8 +--- 2 files changed, 6 insertions(+), 3 deletions(-) diff --git a/configure.ac b/configure.ac index adc3dc1..e637f42

[sheepdog] download vdi via object storage

Does it possible to create vdi for vm and after some time download it via swift like sheepdog object storage? How can i navigate link for vdi in this case? -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo

[sheepdog] [PATCH v1] add libqb corosync depends

sometimes libcfg and libcpg does not have libqb dependencies, but libqb need to be present to link with libcpg and libcfg Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru --- configure.ac | 1 + sheep/Makefile.am | 8 2 files changed, 5 insertions(+), 4 deletions(-) diff --git

Re: [sheepdog] Why Sheepdog is talking more CPU ?

local speed can't be possible because sheep needs to write data to each node. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] Why Sheepdog is talking more CPU ?

or replicas) ? -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] [PATCH] Remove me and Yuan from MAINTAINERS

/sheepdog.py L: openstack-...@lists.openstack.org -- 2.1.4 -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog -- Vasiliy Tolstov, e-mail

[sheepdog] original project and forks

sheepdog-ng not created in github like fork, but create new project and push to them? -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] [sheepdog-users] [ANNOUNCE] Sheepdog stable release v0.9.2

2015-06-09 17:47 GMT+03:00 Hitoshi Mitake mitake.hito...@gmail.com: I'm not familiar with libvirt, so I cannot say it works or not, sorry :( Could you try and report? Yes, ok. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https

Re: [sheepdog] [sheepdog-users] [ANNOUNCE] Sheepdog stable release v0.9.2

2015-06-08 16:38 GMT+03:00 Hitoshi Mitake mitake.hito...@gmail.com: As Fabian pointed, live migration will not work. But in case of --unsafe flag does it work? -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman

Re: [sheepdog] [sheepdog-users] [ANNOUNCE] Sheepdog stable release v0.9.2

use vdi locking and on which does not? For example if i use sheepdog for vm disks and runs vps on compute nodes, does i need vdi locking? And if i enable vdi locking what i miss or what not works after that? -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog

Re: [sheepdog] [sheepdog-users] [ANNOUNCE] Sheepdog stable release v0.9.2

when control server can't connect to compute nodes and think that it die, try to run vps on another node and i have split-brain =() -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] A radical rethinking on struct vdi_state

Per volume redundancy useful for hostings or openstack based solutions. So users may need to pay for stability or increase speed.. 14 мая 2015 г. 13:21 пользователь Valerio Pachera siri...@gmail.com написал: IMHO, per volume redundancy, is something I never felt the need for. -- sheepdog

Re: [sheepdog] [PATCH v2] fix systemd service to run sheep in foreground and with defaults

2015-04-14 5:59 GMT+03:00 Liu Yuan namei.u...@gmail.com: Hi Alexander, Could you please give this patch a review? Thanks, Yuan Any news? -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru jabber: v...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org

[sheepdog] [PATCH v2] fix systemd service to run sheep in foreground and with defaults

sheepdog now able to run in foreground, so use this in systemd service, also remove running shell script to get defaults, because it provided in service file now. Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru --- script/sheepdog.service.in | 5 ++--- 1 file changed, 2 insertions(+), 3

Re: [sheepdog] [PATCH v5] try to set RLIMIT_NOFILE to recommended value

2015-04-07 19:02 GMT+03:00 Hitoshi Mitake mitake.hito...@gmail.com: You need to remove above curly braces, and + +if (r.rlim_cur SD_RLIM_NOFILE) { this check should be skipped if the above getrlimit() fails. Thanks, i'm send new version. -- Vasiliy Tolstov, e-mail: v.tols

[sheepdog] [PATCH v1] fix systemd service to run sheep in foreground and with defaults

sheepdog now able to run in foreground, so use this in systemd service, also remove running shell script to get defaults, because it provided in service file now. Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru --- script/sheepdog.service.in | 6 +++--- 1 file changed, 3 insertions(+), 3

Re: [sheepdog] [PATCH v1] fix systemd service to run sheep in foreground and with defaults

to check. So this is sane defaults to immediately run sheepdog without additional config. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru jabber: v...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

[sheepdog] [PATCH v7] try to set RLIMIT_NOFILE to recommended value

Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru --- sheep/sheep.c | 26 +++-- 1 file changed, 14 insertions(+), 12 deletions(-) diff --git a/sheep/sheep.c b/sheep/sheep.c index 86444b0..4f6a662 100644 --- a/sheep/sheep.c +++ b/sheep/sheep.c @@ -526,22 +526,24 @@ static

Re: [sheepdog] [PATCH v6] try to set RLIMIT_NOFILE to recommended value

2015-04-08 9:18 GMT+03:00 Liu Yuan namei.u...@gmail.com: %m should be appended to show string of errno. +sd_warn(increase nofile via sysctl fs.nr_open); And I'd suggest s/sd_warn/sd_err as mentioned before. Thanks, resend new patch. -- Vasiliy Tolstov, e-mail

Re: [sheepdog] [PATCH v7] try to set RLIMIT_NOFILE to recommended value

2015-04-08 9:46 GMT+03:00 Liu Yuan namei.u...@gmail.com: Applied after fixing style problem found by script/checkpatch.pl. Thanks, Yuan thanks -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru jabber: v...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org

Re: [sheepdog] [PATCH v5] try to set RLIMIT_NOFILE to recommended value

Ping, or something i need to modify to create acceptable patch? 2015-03-26 9:59 GMT+03:00 Vasiliy Tolstov v.tols...@selfip.ru: Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru --- sheep/sheep.c | 29 - 1 file changed, 16 insertions(+), 13 deletions(-) diff

Re: [sheepdog] [PATCH 0/9] introduce libsheepdog.so

2015-04-03 6:49 GMT+03:00 Hitoshi Mitake mitake.hito...@lab.ntt.co.jp: What is the motivation of this shared library? This is good for example to implement own object storage based on sheepdog. For example i can use Go to write server and use this library to store objects. -- Vasiliy Tolstov

[sheepdog] [PATCH v5] try to set RLIMIT_NOFILE to recommended value

Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru --- sheep/sheep.c | 29 - 1 file changed, 16 insertions(+), 13 deletions(-) diff --git a/sheep/sheep.c b/sheep/sheep.c index fc53ec9..3280580 100644 --- a/sheep/sheep.c +++ b/sheep/sheep.c @@ -524,23 +524,26 @@ static

Re: [sheepdog] [PATCH v4] try to set RLIMIT_NOFILE to recommended value

2015-03-26 4:34 GMT+03:00 Liu Yuan namei.u...@gmail.com: sd_err(Failed to set open files limit to suggested %lu, %m, r.rlim_cur) is better so we can see why we fail to setrlimit. Other parts looks good to me. Thanks, Yuan Thanks, i send new patch version. -- Vasiliy Tolstov, e-mail

Re: [sheepdog] [PATCH v2] add foreground option

think need to all daemons , but design =). -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru jabber: v...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

Re: [sheepdog] [PATCH v4] try to set RLIMIT_NOFILE to recommended value

in /etc... Yes, you need sysctl fs.nr_open rise, this is not sheepdog job i think. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru jabber: v...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

[sheepdog] [PATCH v3] add foreground option

This option mainly needed for systemd that does not like forking services. Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru --- sheep/sheep.c | 10 +- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/sheep/sheep.c b/sheep/sheep.c index fc53ec9..86444b0 100644 --- a/sheep

Re: [sheepdog] [PATCH v2] add foreground option

2015-03-25 18:05 GMT+03:00 Hitoshi Mitake mitake.hito...@gmail.com: Seems reasonable. Could you add a comment for describing the new option is mainly for systemd, Vasiliy? Yes, thanks. I'm send v3 version. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru jabber: v...@selfip.ru -- sheepdog

Re: [sheepdog] [PATCH v2] add foreground option

2015-03-18 18:07 GMT+03:00 Vasiliy Tolstov v.tols...@selfip.ru: Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru Does it acceptable? -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru jabber: v...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman

[sheepdog] node identifier in cluster and data placement

when node join to cluster? Or it drop they data and resync? -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru jabber: v...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

[sheepdog] [PATCH v2] add foreground option

Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru --- sheep/sheep.c | 10 +- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/sheep/sheep.c b/sheep/sheep.c index fc53ec9..86444b0 100644 --- a/sheep/sheep.c +++ b/sheep/sheep.c @@ -134,6 +134,7 @@ static struct sd_option

Re: [sheepdog] [PATCH v1] add foreground option

2015-03-17 22:05 GMT+03:00 Alexander Guy alexan...@andern.org: Hey Vasiliy, It doesn't look like you tested this at all. Yes, thanks. I fix patch to more idiomatic way. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru jabber: v...@selfip.ru -- sheepdog mailing list sheepdog

[sheepdog] [PATCH v4] try to set RLIMIT_NOFILE to recommended value

Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru --- sheep/sheep.c | 29 + 1 file changed, 17 insertions(+), 12 deletions(-) diff --git a/sheep/sheep.c b/sheep/sheep.c index fc53ec9..cad8293 100644 --- a/sheep/sheep.c +++ b/sheep/sheep.c @@ -523,19 +523,24 @@ static

[sheepdog] [PATCH v3] try to set RLIMIT_NOFILE to recommended value

Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru --- sheep/sheep.c | 34 -- 1 file changed, 20 insertions(+), 14 deletions(-) diff --git a/sheep/sheep.c b/sheep/sheep.c index fc53ec9..45d364f 100644 --- a/sheep/sheep.c +++ b/sheep/sheep.c @@ -522,21 +522,27

Re: [sheepdog] [PATCH v1] try to set RLIMIT_NOFILE ro recommended value

prlimit sets rlimits to currently running process and it children, but its linux specific. Does this ok for sheepdog? I don't want setrlimit because it sets values to global space. 2015-03-16 19:30 GMT+03:00 Vasiliy Tolstov v.tols...@selfip.ru: Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru

Re: [sheepdog] rlimit nofile

2015-03-16 5:44 GMT+03:00 Liu Yuan namei.u...@gmail.com: Thanks, we should give it a try, could you post a patch for it and verify it can work? I send draft patch, because i think it not accepted for the first time =). -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru jabber: v...@selfip.ru

[sheepdog] [PATCH v1] try to set RLIMIT_NOFILE ro recommended value

Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru --- sheep/sheep.c | 63 +++ 1 file changed, 42 insertions(+), 21 deletions(-) diff --git a/sheep/sheep.c b/sheep/sheep.c index a1028a2..13cc927 100644 --- a/sheep/sheep.c +++ b/sheep

Re: [sheepdog] [PATCH v1] try to set RLIMIT_NOFILE ro recommended value

-03-16 19:30 GMT+03:00 Vasiliy Tolstov v.tols...@selfip.ru: Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru --- sheep/sheep.c | 63 +++ 1 file changed, 42 insertions(+), 21 deletions(-) diff --git a/sheep/sheep.c b/sheep

[sheepdog] [PATCH v2] try to set RLIMIT_NOFILE to recommended value

Signed-off-by: Vasiliy Tolstov v.tols...@selfip.ru --- sheep/sheep.c | 31 ++- 1 file changed, 18 insertions(+), 13 deletions(-) diff --git a/sheep/sheep.c b/sheep/sheep.c index fc53ec9..4a93825 100644 --- a/sheep/sheep.c +++ b/sheep/sheep.c @@ -522,20 +522,25

[sheepdog] rlimit nofile

Why sheepdog complains about maximum open files limit is too small and not increase it via http://man7.org/linux/man-pages/man2/getrlimit.2.html ? -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru jabber: v...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org

Re: [sheepdog] writing cluster driver

. Thanks, i'm try. I assume etcd driver is similar to zookeeper, once you grasp local.c, the next is to take a look at zookeeper.c white write etcd driver. Yes, i already think about this. Thanks again. -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru jabber: v...@selfip.ru -- sheepdog mailing

[sheepdog] writing cluster driver

of this function executed? Thanks -- Vasiliy Tolstov, e-mail: v.tols...@selfip.ru jabber: v...@selfip.ru -- sheepdog mailing list sheepdog@lists.wpkg.org https://lists.wpkg.org/mailman/listinfo/sheepdog

  1   2   >