Re: [Gluster-users] small files performance
Hi, Parallel-readdir is an experimental feature for 3.10, can you disable performance.parallel-readdir option and see if the files are visible? Does the unmount-mount help? Also If you want to use parallel-readdir in production please use 3.11 or greater. Regards, Poornima - Original Message - > From: "Alastair Neil"> To: "gluster-users" > Sent: Wednesday, October 11, 2017 3:29:10 AM > Subject: Re: [Gluster-users] small files performance > I just tried setting: > performance.parallel-readdir on > features.cache-invalidation on > features.cache-invalidation-timeout 600 > performance.stat-prefetch > performance.cache-invalidation > performance.md-cache-timeout 600 > network.inode-lru-limit 5 > performance.cache-invalidation on > and clients could not see their files with ls when accessing via a fuse > mount. The files and directories were there, however, if you accessed them > directly. Server are 3.10.5 and the clients are 3.10 and 3.12. > Any ideas? > On 10 October 2017 at 10:53, Gandalf Corvotempesta < > gandalf.corvotempe...@gmail.com > wrote: > > 2017-10-10 8:25 GMT+02:00 Karan Sandha < ksan...@redhat.com > : > > > > Hi Gandalf, > > > > > > We have multiple tuning to do for small-files which decrease the time for > > > negative lookups , meta-data caching, parallel readdir. Bumping the > > > server > > > and client event threads will help you out in increasing the small file > > > performance. > > > > > > gluster v set group metadata-cache > > > > > > gluster v set group nl-cache > > > > > > gluster v set performance.parallel-readdir on (Note : readdir > > > should be on) > > > > > This is what i'm getting with suggested parameters. > > > I'm running "fio" from a mounted gluster client: > > > 172.16.0.12:/gv0 on /mnt2 type fuse.glusterfs > > (rw,relatime,user_id=0,group_id=0,default_permissions,allow_other,max_read=131072) > > > # fio --ioengine=libaio --filename=fio.test --size=256M --direct=1 > > --rw=randrw --refill_buffers --norandommap --bs=8k --rwmixread=70 > > --iodepth=16 --numjobs=16 --runtime=60 --group_reporting --name=fio-test > > > fio-test: (g=0): rw=randrw, bs=8K-8K/8K-8K/8K-8K, ioengine=libaio, > > iodepth=16 > > > ... > > > fio-2.16 > > > Starting 16 processes > > > fio-test: Laying out IO file(s) (1 file(s) / 256MB) > > > Jobs: 14 (f=13): [m(5),_(1),m(8),f(1),_(1)] [33.9% done] [1000KB/440KB/0KB > > /s] [125/55/0 iops] [eta 01m:59s] > > > fio-test: (groupid=0, jobs=16): err= 0: pid=2051: Tue Oct 10 16:51:46 2017 > > > read : io=43392KB, bw=733103B/s, iops=89, runt= 60610msec > > > slat (usec): min=14, max=1992.5K, avg=177873.67, stdev=382294.06 > > > clat (usec): min=768, max=6016.8K, avg=1871390.57, stdev=1082220.06 > > > lat (usec): min=872, max=6630.6K, avg=2049264.23, stdev=1158405.41 > > > clat percentiles (msec): > > > | 1.00th=[ 20], 5.00th=[ 208], 10.00th=[ 457], 20.00th=[ 873], > > > | 30.00th=[ 1237], 40.00th=[ 1516], 50.00th=[ 1795], 60.00th=[ 2073], > > > | 70.00th=[ 2442], 80.00th=[ 2835], 90.00th=[ 3326], 95.00th=[ 3785], > > > | 99.00th=[ 4555], 99.50th=[ 4948], 99.90th=[ 5211], 99.95th=[ 5800], > > > | 99.99th=[ 5997] > > > write: io=18856KB, bw=318570B/s, iops=38, runt= 60610msec > > > slat (usec): min=17, max=3428, avg=212.62, stdev=287.88 > > > clat (usec): min=59, max=6015.6K, avg=1693729.12, stdev=1003122.83 > > > lat (usec): min=79, max=6015.9K, avg=1693941.74, stdev=1003126.51 > > > clat percentiles (usec): > > > | 1.00th=[ 724], 5.00th=[144384], 10.00th=[403456], 20.00th=[765952], > > > | 30.00th=[1105920], 40.00th=[1368064], 50.00th=[1630208], > > | 60.00th=[1875968], > > > | 70.00th=[2179072], 80.00th=[2572288], 90.00th=[3031040], > > | 95.00th=[3489792], > > > | 99.00th=[4227072], 99.50th=[4423680], 99.90th=[4751360], > > | 99.95th=[5210112], > > > | 99.99th=[5996544] > > > lat (usec) : 100=0.15%, 250=0.05%, 500=0.06%, 750=0.09%, 1000=0.05% > > > lat (msec) : 2=0.28%, 4=0.09%, 10=0.15%, 20=0.39%, 50=1.81% > > > lat (msec) : 100=1.02%, 250=1.63%, 500=5.59%, 750=6.03%, 1000=7.31% > > > lat (msec) : 2000=35.61%, >=2000=39.67% > > > cpu : usr=0.01%, sys=0.01%, ctx=8218, majf=11, minf=295 > > > IO depths : 1=0.2%, 2=0.4%, 4=0.8%, 8=1.6%, 16=96.9%, 32=0.0%, >=64=0.0% > > > 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=99.8%, 8=0.0%, 16=0.2%, 32=0.0%, 64=0.0%, >=64=0.0% > > > issued : total=r=5424/w=2357/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=16 > > > Run status group 0 (all jobs): > > > READ: io=43392KB, aggrb=715KB/s, minb=715KB/s, maxb=715KB/s, > > mint=60610msec, > > maxt=60610msec > > > WRITE: io=18856KB, aggrb=311KB/s, minb=311KB/s, maxb=311KB/s, > > mint=60610msec, > > maxt=60610msec > > > ___ > > > Gluster-users mailing list > > >
Re: [Gluster-users] data corruption - any update?
On 11 October 2017 at 22:21,wrote: > > corruption happens only in this cases: > > > > - volume with shard enabled > > AND > > - rebalance operation > > > > I believe so > > > So, what If I have to replace a failed brick/disks ? Will this trigger > > a rebalance and then corruption? > > > > rebalance, is only needed when you have to expend a volume, ie by > > adding more bricks ? > > That's correct, replacing a brick shouldn't cause corruption, I've done > it a few times without any problems. As long as you don't expand the > cluster, you are fine. > > Basically you can add or remove replicas all you want, but you can't add > new replica sets. > Or remove a replica set. An add-brick will not trigger a rebalance - that needs to be done explicitly. However, a remove-brick will start the rebalance automatically. Regards, Nithya > > ___ > Gluster-users mailing list > Gluster-users@gluster.org > http://lists.gluster.org/mailman/listinfo/gluster-users > ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users
[Gluster-users] how does gluster decide which connection to use?
hi everyone I assume such a situation where network segement changes, in a simplest case where one provides a box(a brick) a new faster net iterface. So after boxes have two nics and then bricks gets introduced to them via gluster probe $_newIPs. Ideally @ a developer - how gluster handles such a situation? Will it deterministically decide which connection to use? Would there be a doc/howto on switching/migrating NICs(not by a replacement but by an expansion). I sroogled but failed to find one. many thanks, L. ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users
Re: [Gluster-users] data corruption - any update?
> corruption happens only in this cases: > > - volume with shard enabled > AND > - rebalance operation > I believe so > So, what If I have to replace a failed brick/disks ? Will this trigger > a rebalance and then corruption? > > rebalance, is only needed when you have to expend a volume, ie by > adding more bricks ? That's correct, replacing a brick shouldn't cause corruption, I've done it a few times without any problems. As long as you don't expand the cluster, you are fine. Basically you can add or remove replicas all you want, but you can't add new replica sets. signature.asc Description: Digital signature ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users
Re: [Gluster-users] gluster volume + lvm : recommendation or neccessity ?
2017-10-11 15:37 GMT+02:00 ML: > After some extra reading about LVM snapshots & Gluster, I think I can > conclude it may be a bad idea to use it on big storage bricks. > > I understood that the LVM maximum metadata, used to store the snapshots > data, is about 16GB. > LVM metadata aer used to store changed METADATA, not data. thin-provisioned snapshots usually may grow up to the local unallocated capacity. > So if I have a brick with a volume arount 10TB (for example), daily > snapshots, files changing ~100GB : the LVM snapshot is useless. > > LVM's snapshots doesn't seems to be a good idea with very big LVM > partitions. > > Did I missed something ? Hard to find clear documentation on the subject. > LVM documentation (RH has very good docs available via web) and even the lvcreate man page is OK. not a lightweight read but OK. You need thin-provisioned lvm pools to have snapshots in gluster. > > Le 11/10/2017 à 09:07, Ric Wheeler a écrit : > >> On 10/11/2017 09:50 AM, ML wrote: >> >>> Hi everyone, >>> >>> I've read on the gluster & redhat documentation, that it seems >>> recommended to use XFS over LVM before creating & using gluster volumes. >>> >>> Sources : >>> https://access.redhat.com/documentation/en-US/Red_Hat_Storag >>> e/3/html/Administration_Guide/Formatting_and_Mounting_Bricks.html >>> http://gluster.readthedocs.io/en/latest/Administrator%20Guid >>> e/Setting%20Up%20Volumes/ >>> >>> My point is : do we really need LVM ? >>> For example , on a dedicated server with disks & partitions that will >>> not change of size, it doesn't seems necessary to use LVM. >>> >>> I can't understand clearly wich partitioning strategy would be the best >>> for "static size" hard drives : >>> >>> 1 LVM+XFS partition = multiple gluster volumes >>> or 1 LVM+XFS partition = 1 gluster volume per LVM+XFS partition >>> or 1 XFS partition = multiple gluster volumes >>> or 1 XFS partition = 1 gluster volume per XFS partition >>> >>> What do you use on your servers ? >>> >>> Thanks for your help! :) >>> >>> Quentin >>> >> >> Hi Quentin, >> >> Gluster relies on LVM for snapshots - you won't get those unless you >> deploy on LVM. >> >> Regards, >> Ric >> >> ___ >> Gluster-users mailing list >> Gluster-users@gluster.org >> http://lists.gluster.org/mailman/listinfo/gluster-users >> > > ___ > Gluster-users mailing list > Gluster-users@gluster.org > http://lists.gluster.org/mailman/listinfo/gluster-users > ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users
[Gluster-users] Community Meeting 2017-10-11
We had a quick meeting today, with 2 main topics. We have a new community issue tracker [1], which will be used to track community initiatives. Amye will be sharing more information about this in another email. To co-ordinate people travelling to the Gluster Community Summit better, a spreadsheet [2] has been setup to share information. Apart from the above 2 topics, Shyam shared that he is on the lookout for a partner to manage the 4.0 release. For more information, meeting logs and minutes are available at the links below. [3][4][5] The meeting scheduled to be held on the 25 Oct, is being skipped. A lot of the attendees will be travelling to the Gluster Summit at the time. The next meeting now is scheduled for 8th Nov. See you then. ~kaushal [1]: https://github.com/gluster/community [2]: https://docs.google.com/spreadsheets/d/1Jde-5XNc0q4a8bW8-OmLC2w_jiPg-e53ssR4wanIhFk/edit#gid=0 [3]: Minutes: https://meetbot.fedoraproject.org/gluster-meeting/2017-10-11/gluster_community_meeting_2017-10-11.2017-10-11-15.03.html [4]: Minutes (text): https://meetbot.fedoraproject.org/gluster-meeting/2017-10-11/gluster_community_meeting_2017-10-11.2017-10-11-15.03.txt [5]: Log: https://meetbot.fedoraproject.org/gluster-meeting/2017-10-11/gluster_community_meeting_2017-10-11.2017-10-11-15.03.log.html ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users
Re: [Gluster-users] data corruption - any update?
Just to clarify as i'm planning to put gluster in production (after fixing some issue, but for this I need community help): corruption happens only in this cases: - volume with shard enabled AND - rebalance operation In any other cases, corruption should not happen (or at least is not known to happen) So, what If I have to replace a failed brick/disks ? Will this trigger a rebalance and then corruption? rebalance, is only needed when you have to expend a volume, ie by adding more bricks ? 2017-10-05 13:55 GMT+02:00 Nithya Balachandran: > > > On 4 October 2017 at 23:34, WK wrote: >> >> Just so I know. >> >> Is it correct to assume that this corruption issue is ONLY involved if you >> are doing rebalancing with sharding enabled. >> >> So if I am not doing rebalancing I should be fine? > > > That is correct. > >> >> -bill >> >> >> >> On 10/3/2017 10:30 PM, Krutika Dhananjay wrote: >> >> >> >> On Wed, Oct 4, 2017 at 10:51 AM, Nithya Balachandran >> wrote: >>> >>> >>> >>> On 3 October 2017 at 13:27, Gandalf Corvotempesta >>> wrote: Any update about multiple bugs regarding data corruptions with sharding enabled ? Is 3.12.1 ready to be used in production? >>> >>> >>> Most issues have been fixed but there appears to be one more race for >>> which the patch is being worked on. >>> >>> @Krutika, is that correct? >>> >>> >> >> That is my understanding too, yes, in light of the discussion that >> happened at https://bugzilla.redhat.com/show_bug.cgi?id=1465123 >> >> -Krutika >> >>> >>> Thanks, >>> Nithya >>> >>> >>> ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users >>> >>> >> >> >> >> ___ >> Gluster-users mailing list >> Gluster-users@gluster.org >> http://lists.gluster.org/mailman/listinfo/gluster-users >> >> >> >> ___ >> Gluster-users mailing list >> Gluster-users@gluster.org >> http://lists.gluster.org/mailman/listinfo/gluster-users > > > > ___ > Gluster-users mailing list > Gluster-users@gluster.org > http://lists.gluster.org/mailman/listinfo/gluster-users ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users
Re: [Gluster-users] gluster volume + lvm : recommendation or neccessity ?
LVM is also good if you want to add ssd cache. It is more flexible and easier to manage and expand than bcache. On 11 October 2017 at 04:00, Mohammed Rafi K Cwrote: > > Volumes are aggregation of bricks, so I would consider bricks as a > unique entity here rather than volumes. Taking the constraints from the > blog [1]. > > * All bricks should be carved out from an independent thinly provisioned > logical volume (LV). In other words, no two brick should share a common > LV. More details about thin provisioning and thin provisioned snapshot > can be found here. > * This thinly provisioned LV should only be used for forming a brick. > * Thin pool from which the thin LVs are created should have sufficient > space and also it should have sufficient space for pool metadata. > > You can refer the blog post here [1]. > > [1] : http://rajesh-joseph.blogspot.in/p/gluster-volume-snapshot- > howto.html > > Regards > Rafi KC > > > On 10/11/2017 01:23 PM, ML wrote: > > Thanks Rafi, that's understood now :) > > > > I'm considering to deploy gluster on a 4 x 40 TB bricks, do you think > > it would better to make 1 LVM partition for each Volume I need or to > > make one Big LVM partition and start multiple volumes on it ? > > > > We'll store mostly big files (videos) on this environement. > > > > > > > > > > Le 11/10/2017 à 09:34, Mohammed Rafi K C a écrit : > >> > >> On 10/11/2017 12:20 PM, ML wrote: > >>> Hi everyone, > >>> > >>> I've read on the gluster & redhat documentation, that it seems > >>> recommended to use XFS over LVM before creating & using gluster > >>> volumes. > >>> > >>> Sources : > >>> https://access.redhat.com/documentation/en-US/Red_Hat_ > Storage/3/html/Administration_Guide/Formatting_and_Mounting_Bricks.html > >>> > >>> > >>> http://gluster.readthedocs.io/en/latest/Administrator% > 20Guide/Setting%20Up%20Volumes/ > >>> > >>> > >>> > >>> My point is : do we really need LVM ? > >> This recommendations was added after gluster-snapshot. Gluster snapshot > >> relays on LVM snapshot. So if you start with out lvm, in future if you > >> want to use snapshot then it would be difficult, hence the > >> recommendation to use xfs on top of lvm. > >> > >> > >> Regards > >> Rafi KC > >> > >>> For example , on a dedicated server with disks & partitions that will > >>> not change of size, it doesn't seems necessary to use LVM. > >>> > >>> I can't understand clearly wich partitioning strategy would be the > >>> best for "static size" hard drives : > >>> > >>> 1 LVM+XFS partition = multiple gluster volumes > >>> or 1 LVM+XFS partition = 1 gluster volume per LVM+XFS partition > >>> or 1 XFS partition = multiple gluster volumes > >>> or 1 XFS partition = 1 gluster volume per XFS partition > >>> > >>> What do you use on your servers ? > >>> > >>> Thanks for your help! :) > >>> > >>> Quentin > >>> > >>> > >>> ___ > >>> Gluster-users mailing list > >>> Gluster-users@gluster.org > >>> http://lists.gluster.org/mailman/listinfo/gluster-users > > > > ___ > Gluster-users mailing list > Gluster-users@gluster.org > http://lists.gluster.org/mailman/listinfo/gluster-users > ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users
Re: [Gluster-users] Gluster CLI Feedback
Hi, I have only a feedback to point 4/5. Despite using http://docs.ansible.com/ansible/latest/gluster_volume_module.html for gluster management I see the operation of replacing a server with a new hardware while keeping the same IP number poorly documented. Maybe I was just unlucky in my search for the documentation, but the best source I found is https://support.rackspace.com/how-to/recover-from-a-failed-server-in-a-glusterfs-array/ Basically I need to: shut down the old server, perform a clean OS installation on the new server, and make the new server to appear disguised as the old peer, while limiting the bandwidth used by the volume sync. Cheers, Marcin On Wed, Oct 11, 2017 at 11:08 AM, Nithya Balachandranwrote: > Hi, > > As part of our initiative to improve Gluster usability, we would like > feedback on the current Gluster CLI. Gluster 4.0 upstream development is > currently in progress and it is an ideal time to consider CLI changes. > Answers to the following would be appreciated: > >1. How often do you use the Gluster CLI? Is it a preferred method to >manage Gluster? >2. What operations do you commonly perform using the CLI? >3. How intuitive/easy to use do you find the CLI ? >4. Is the help/information displayed sufficient? >5. Are there operations that are difficult to perform? > > Regards, > Nithya > > ___ > Gluster-users mailing list > Gluster-users@gluster.org > http://lists.gluster.org/mailman/listinfo/gluster-users > ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users
Re: [Gluster-users] gluster volume + lvm : recommendation or neccessity ?
After some extra reading about LVM snapshots & Gluster, I think I can conclude it may be a bad idea to use it on big storage bricks. I understood that the LVM maximum metadata, used to store the snapshots data, is about 16GB. So if I have a brick with a volume arount 10TB (for example), daily snapshots, files changing ~100GB : the LVM snapshot is useless. LVM's snapshots doesn't seems to be a good idea with very big LVM partitions. Did I missed something ? Hard to find clear documentation on the subject. ++ Quentin Le 11/10/2017 à 09:07, Ric Wheeler a écrit : On 10/11/2017 09:50 AM, ML wrote: Hi everyone, I've read on the gluster & redhat documentation, that it seems recommended to use XFS over LVM before creating & using gluster volumes. Sources : https://access.redhat.com/documentation/en-US/Red_Hat_Storage/3/html/Administration_Guide/Formatting_and_Mounting_Bricks.html http://gluster.readthedocs.io/en/latest/Administrator%20Guide/Setting%20Up%20Volumes/ My point is : do we really need LVM ? For example , on a dedicated server with disks & partitions that will not change of size, it doesn't seems necessary to use LVM. I can't understand clearly wich partitioning strategy would be the best for "static size" hard drives : 1 LVM+XFS partition = multiple gluster volumes or 1 LVM+XFS partition = 1 gluster volume per LVM+XFS partition or 1 XFS partition = multiple gluster volumes or 1 XFS partition = 1 gluster volume per XFS partition What do you use on your servers ? Thanks for your help! :) Quentin Hi Quentin, Gluster relies on LVM for snapshots - you won't get those unless you deploy on LVM. Regards, Ric ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users
[Gluster-users] iozone results
I'm testing iozone inside a VM booted from a gluster volume. By looking at network traffic on the host (the one connected to the gluster storage) I can see that a simple iozone -w -c -e -i 0 -+n -C -r 64k -s 1g -t 1 -F /tmp/gluster.ioz will make about 1200mbit/s on a bonded dual gigabit nic (probably, with a bad bonding mode configured) fio returns about 5kB/s, that are 40 kbps. As I'm using replica 3, the host has to write to 3 storage server, thus: 40*3 = 120 kbps If I understood properly, i'm able to reach about 120 kpbs on network side, with sequential writes, right ? Why a simple "dd" will return only 30MB/s ? ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users
[Gluster-users] Gluster CLI Feedback
Hi, As part of our initiative to improve Gluster usability, we would like feedback on the current Gluster CLI. Gluster 4.0 upstream development is currently in progress and it is an ideal time to consider CLI changes. Answers to the following would be appreciated: 1. How often do you use the Gluster CLI? Is it a preferred method to manage Gluster? 2. What operations do you commonly perform using the CLI? 3. How intuitive/easy to use do you find the CLI ? 4. Is the help/information displayed sufficient? 5. Are there operations that are difficult to perform? Regards, Nithya ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users
Re: [Gluster-users] gluster volume + lvm : recommendation or neccessity ?
Volumes are aggregation of bricks, so I would consider bricks as a unique entity here rather than volumes. Taking the constraints from the blog [1]. * All bricks should be carved out from an independent thinly provisioned logical volume (LV). In other words, no two brick should share a common LV. More details about thin provisioning and thin provisioned snapshot can be found here. * This thinly provisioned LV should only be used for forming a brick. * Thin pool from which the thin LVs are created should have sufficient space and also it should have sufficient space for pool metadata. You can refer the blog post here [1]. [1] : http://rajesh-joseph.blogspot.in/p/gluster-volume-snapshot-howto.html Regards Rafi KC On 10/11/2017 01:23 PM, ML wrote: > Thanks Rafi, that's understood now :) > > I'm considering to deploy gluster on a 4 x 40 TB bricks, do you think > it would better to make 1 LVM partition for each Volume I need or to > make one Big LVM partition and start multiple volumes on it ? > > We'll store mostly big files (videos) on this environement. > > > > > Le 11/10/2017 à 09:34, Mohammed Rafi K C a écrit : >> >> On 10/11/2017 12:20 PM, ML wrote: >>> Hi everyone, >>> >>> I've read on the gluster & redhat documentation, that it seems >>> recommended to use XFS over LVM before creating & using gluster >>> volumes. >>> >>> Sources : >>> https://access.redhat.com/documentation/en-US/Red_Hat_Storage/3/html/Administration_Guide/Formatting_and_Mounting_Bricks.html >>> >>> >>> http://gluster.readthedocs.io/en/latest/Administrator%20Guide/Setting%20Up%20Volumes/ >>> >>> >>> >>> My point is : do we really need LVM ? >> This recommendations was added after gluster-snapshot. Gluster snapshot >> relays on LVM snapshot. So if you start with out lvm, in future if you >> want to use snapshot then it would be difficult, hence the >> recommendation to use xfs on top of lvm. >> >> >> Regards >> Rafi KC >> >>> For example , on a dedicated server with disks & partitions that will >>> not change of size, it doesn't seems necessary to use LVM. >>> >>> I can't understand clearly wich partitioning strategy would be the >>> best for "static size" hard drives : >>> >>> 1 LVM+XFS partition = multiple gluster volumes >>> or 1 LVM+XFS partition = 1 gluster volume per LVM+XFS partition >>> or 1 XFS partition = multiple gluster volumes >>> or 1 XFS partition = 1 gluster volume per XFS partition >>> >>> What do you use on your servers ? >>> >>> Thanks for your help! :) >>> >>> Quentin >>> >>> >>> ___ >>> Gluster-users mailing list >>> Gluster-users@gluster.org >>> http://lists.gluster.org/mailman/listinfo/gluster-users > ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users
Re: [Gluster-users] gluster volume + lvm : recommendation or neccessity ?
Thanks Rafi, that's understood now :) I'm considering to deploy gluster on a 4 x 40 TB bricks, do you think it would better to make 1 LVM partition for each Volume I need or to make one Big LVM partition and start multiple volumes on it ? We'll store mostly big files (videos) on this environement. Le 11/10/2017 à 09:34, Mohammed Rafi K C a écrit : On 10/11/2017 12:20 PM, ML wrote: Hi everyone, I've read on the gluster & redhat documentation, that it seems recommended to use XFS over LVM before creating & using gluster volumes. Sources : https://access.redhat.com/documentation/en-US/Red_Hat_Storage/3/html/Administration_Guide/Formatting_and_Mounting_Bricks.html http://gluster.readthedocs.io/en/latest/Administrator%20Guide/Setting%20Up%20Volumes/ My point is : do we really need LVM ? This recommendations was added after gluster-snapshot. Gluster snapshot relays on LVM snapshot. So if you start with out lvm, in future if you want to use snapshot then it would be difficult, hence the recommendation to use xfs on top of lvm. Regards Rafi KC For example , on a dedicated server with disks & partitions that will not change of size, it doesn't seems necessary to use LVM. I can't understand clearly wich partitioning strategy would be the best for "static size" hard drives : 1 LVM+XFS partition = multiple gluster volumes or 1 LVM+XFS partition = 1 gluster volume per LVM+XFS partition or 1 XFS partition = multiple gluster volumes or 1 XFS partition = 1 gluster volume per XFS partition What do you use on your servers ? Thanks for your help! :) Quentin ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users
Re: [Gluster-users] gluster volume + lvm : recommendation or neccessity ?
On 10/11/2017 09:50 AM, ML wrote: Hi everyone, I've read on the gluster & redhat documentation, that it seems recommended to use XFS over LVM before creating & using gluster volumes. Sources : https://access.redhat.com/documentation/en-US/Red_Hat_Storage/3/html/Administration_Guide/Formatting_and_Mounting_Bricks.html http://gluster.readthedocs.io/en/latest/Administrator%20Guide/Setting%20Up%20Volumes/ My point is : do we really need LVM ? For example , on a dedicated server with disks & partitions that will not change of size, it doesn't seems necessary to use LVM. I can't understand clearly wich partitioning strategy would be the best for "static size" hard drives : 1 LVM+XFS partition = multiple gluster volumes or 1 LVM+XFS partition = 1 gluster volume per LVM+XFS partition or 1 XFS partition = multiple gluster volumes or 1 XFS partition = 1 gluster volume per XFS partition What do you use on your servers ? Thanks for your help! :) Quentin Hi Quentin, Gluster relies on LVM for snapshots - you won't get those unless you deploy on LVM. Regards, Ric ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users
Re: [Gluster-users] gluster volume + lvm : recommendation or neccessity ?
just had an answer here for those interrested : https://github.com/gluster/glusterdocs/issues/218 Le 11/10/2017 à 08:50, ML a écrit : Hi everyone, I've read on the gluster & redhat documentation, that it seems recommended to use XFS over LVM before creating & using gluster volumes. Sources : https://access.redhat.com/documentation/en-US/Red_Hat_Storage/3/html/Administration_Guide/Formatting_and_Mounting_Bricks.html http://gluster.readthedocs.io/en/latest/Administrator%20Guide/Setting%20Up%20Volumes/ My point is : do we really need LVM ? For example , on a dedicated server with disks & partitions that will not change of size, it doesn't seems necessary to use LVM. I can't understand clearly wich partitioning strategy would be the best for "static size" hard drives : 1 LVM+XFS partition = multiple gluster volumes or 1 LVM+XFS partition = 1 gluster volume per LVM+XFS partition or 1 XFS partition = multiple gluster volumes or 1 XFS partition = 1 gluster volume per XFS partition What do you use on your servers ? Thanks for your help! :) Quentin ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users
Re: [Gluster-users] gluster volume + lvm : recommendation or neccessity ?
On 10/11/2017 12:20 PM, ML wrote: > Hi everyone, > > I've read on the gluster & redhat documentation, that it seems > recommended to use XFS over LVM before creating & using gluster volumes. > > Sources : > https://access.redhat.com/documentation/en-US/Red_Hat_Storage/3/html/Administration_Guide/Formatting_and_Mounting_Bricks.html > > http://gluster.readthedocs.io/en/latest/Administrator%20Guide/Setting%20Up%20Volumes/ > > > My point is : do we really need LVM ? This recommendations was added after gluster-snapshot. Gluster snapshot relays on LVM snapshot. So if you start with out lvm, in future if you want to use snapshot then it would be difficult, hence the recommendation to use xfs on top of lvm. Regards Rafi KC > For example , on a dedicated server with disks & partitions that will > not change of size, it doesn't seems necessary to use LVM. > > I can't understand clearly wich partitioning strategy would be the > best for "static size" hard drives : > > 1 LVM+XFS partition = multiple gluster volumes > or 1 LVM+XFS partition = 1 gluster volume per LVM+XFS partition > or 1 XFS partition = multiple gluster volumes > or 1 XFS partition = 1 gluster volume per XFS partition > > What do you use on your servers ? > > Thanks for your help! :) > > Quentin > > > ___ > Gluster-users mailing list > Gluster-users@gluster.org > http://lists.gluster.org/mailman/listinfo/gluster-users ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users
[Gluster-users] gluster volume + lvm : recommendation or neccessity ?
Hi everyone, I've read on the gluster & redhat documentation, that it seems recommended to use XFS over LVM before creating & using gluster volumes. Sources : https://access.redhat.com/documentation/en-US/Red_Hat_Storage/3/html/Administration_Guide/Formatting_and_Mounting_Bricks.html http://gluster.readthedocs.io/en/latest/Administrator%20Guide/Setting%20Up%20Volumes/ My point is : do we really need LVM ? For example , on a dedicated server with disks & partitions that will not change of size, it doesn't seems necessary to use LVM. I can't understand clearly wich partitioning strategy would be the best for "static size" hard drives : 1 LVM+XFS partition = multiple gluster volumes or 1 LVM+XFS partition = 1 gluster volume per LVM+XFS partition or 1 XFS partition = multiple gluster volumes or 1 XFS partition = 1 gluster volume per XFS partition What do you use on your servers ? Thanks for your help! :) Quentin ___ Gluster-users mailing list Gluster-users@gluster.org http://lists.gluster.org/mailman/listinfo/gluster-users