[lustre-discuss] changing inode size on MDT

Hebenstreit, Michael michael.hebenstreit at intel.com
Thu Oct 3 07:12:30 PDT 2019


Please note the differences between inodes on the ZFS and inodes on the mdt lustre. In the previous incarnation the file system run out of ionodes as reported on the Lustre, even though the mdt was only half filled and zfs backend still reported free inodes.

From: Shaun Tancheff <stancheff at cray.com>
Sent: Thursday, October 03, 2019 05:41
To: Degremont, Aurelien <degremoa at amazon.com>; Hebenstreit, Michael <michael.hebenstreit at intel.com>; Andreas Dilger <adilger at whamcloud.com>
Cc: lustre-discuss at lists.lustre.org
Subject: Re: [lustre-discuss] changing inode size on MDT

Hi,

A little pedantic but for ‘inodes’ don’t exist in a zfs pool per-se. So the code which attempts to report the number of inodes used/available guesses based on the average per-object utilization rate. If you have a many large files your reported number of inodes goes down faster than if you have a many small files.

I think the primary take away is that for a zfs backend you will run out of _inodes_ at the same time as you run out of _space_, there simply is no distinction.

From: lustre-discuss <lustre-discuss-bounces at lists.lustre.org<mailto:lustre-discuss-bounces at lists.lustre.org>> on behalf of "Degremont, Aurelien" <degremoa at amazon.com<mailto:degremoa at amazon.com>>
Date: Thursday, October 3, 2019 at 6:11 PM
To: "Hebenstreit, Michael" <michael.hebenstreit at intel.com<mailto:michael.hebenstreit at intel.com>>, Andreas Dilger <adilger at whamcloud.com<mailto:adilger at whamcloud.com>>
Cc: "lustre-discuss at lists.lustre.org<mailto:lustre-discuss at lists.lustre.org>" <lustre-discuss at lists.lustre.org<mailto:lustre-discuss at lists.lustre.org>>
Subject: Re: [lustre-discuss] changing inode size on MDT

As Andreas said "it is not relevant for ZFS since ZFS dynamically allocates inodes and blocks as needed"

"as needed" is the important part. In your example, your MDT is almost empty, so 17G inodes for an empty MDT seems pretty sufficient.
As you will create new files and use these inodes, you will see the total number of inodes change.

But I don't what is the maximum number of inodes you can estimate for a 8.3 TiB MDT…

De : lustre-discuss <lustre-discuss-bounces at lists.lustre.org<mailto:lustre-discuss-bounces at lists.lustre.org>> au nom de "Hebenstreit, Michael" <michael.hebenstreit at intel.com<mailto:michael.hebenstreit at intel.com>>
Date : jeudi 3 octobre 2019 à 13:04
À : Andreas Dilger <adilger at whamcloud.com<mailto:adilger at whamcloud.com>>
Cc : "lustre-discuss at lists.lustre.org<mailto:lustre-discuss at lists.lustre.org>" <lustre-discuss at lists.lustre.org<mailto:lustre-discuss at lists.lustre.org>>
Objet : Re: [lustre-discuss] changing inode size on MDT

So you are saying on a zfs based Lustre there is no way to increase the number of available inodes? I have 8TB MDT with roughly 17G inodes

[root at elfsa1m1 ~]# df -h
Filesystem       Size  Used Avail Use% Mounted on
mdt0000          8.3T  256K  8.3T   1% /mdt0000

[root at elfsa1m1 ~]# df -i
Filesystem           Inodes  IUsed       IFree IUse% Mounted on
mdt0000         17678817874      6 17678817868    1% /mdt0000

Formating under Lustre 2.10.8

mkfs.lustre --mdt --backfstype=zfs --fsname=lfsarc01 --index=0 --mgsnid="36.101.92.22 at tcp<mailto:36.101.92.22 at tcp>" --reformat mdt0000/mdt0000

this translates to only 948M inodes on the Lustre FS.

[root at elfsa1m1 ~]# df -i
Filesystem           Inodes  IUsed       IFree IUse% Mounted on
mdt0000         17678817874      6 17678817868    1% /mdt0000
mdt0000/mdt0000   948016092    263   948015829    1% /lfs/lfsarc01/mdt

[root at elfsa1m1 ~]# df -h
Filesystem       Size  Used Avail Use% Mounted on
mdt0000          8.3T  256K  8.3T   1% /mdt0000
mdt0000/mdt0000  8.2T   24M  8.2T   1% /lfs/lfsarc01/mdt

and there is no reasonable option to provide more file entries except for adding another MDT?

Thanks
Michael

From: Andreas Dilger <adilger at whamcloud.com<mailto:adilger at whamcloud.com>>
Sent: Wednesday, October 02, 2019 18:49
To: Hebenstreit, Michael <michael.hebenstreit at intel.com<mailto:michael.hebenstreit at intel.com>>
Cc: Mohr Jr, Richard Frank <rmohr at utk.edu<mailto:rmohr at utk.edu>>; lustre-discuss at lists.lustre.org<mailto:lustre-discuss at lists.lustre.org>
Subject: Re: [lustre-discuss] changing inode size on MDT

There are several confusing/misleading comments on this thread that need to be clarified...

On Oct 2, 2019, at 13:45, Hebenstreit, Michael <michael.hebenstreit at intel.com<mailto:michael.hebenstreit at intel.com>> wrote:

http://wiki.lustre.org/Lustre_Tuning#Number_of_Inodes_for_MDS

Note that I've updated this page to reflect current defaults.  The Lustre Operations Manual has a much better description of these parameters.


and I'd like to use --mkfsoptions='-i 1024' to have more inodes in the MDT. We already run out of inodes on that FS (probably due to an ZFS bug in early IEEL version) - so I'd like to increase #inodes if possible.

The "-i 1024" option (bytes-per-inode ratio) is only needed for ldiskfs since it statically allocates the inodes at mkfs time, it is not relevant for ZFS since ZFS dynamically allocates inodes and blocks as needed.

On Oct 2, 2019, at 14:00, Colin Faber <cfaber at gmail.com<mailto:cfaber at gmail.com>> wrote:
With 1K inodes you won't have space to accommodate new features, IIRC the current minimal limit on modern lustre is 2K now. If you're running out of MDT space you might consider DNE and multiple MDT's to accommodate that larger name space.

To clarify, since Lustre 2.10 any new ldiskfs MDT will allocate 1024 bytes for the inode itself (-I 1024).  That allows enough space *within* the inode to efficiently store xattrs for more complex layouts (PFL, FLR, DoM).  If xattrs do not fit inside the inode itself then they will be stored in an external 4KB inode block.

The MDT is formatted with a bytes-per-inode *ratio* of 2.5KB, which means (approximately) one inode will be created for every 2.5kB of the total MDT size.  That 2.5KB of space includes the 1KB for the inode itself, plus space for a directory entry (or multiple if hard-linked), extra xattrs, the journal (up to 4GB for large MDTs), Lustre recovery logs, ChangeLogs, etc.  Each directory inode will have at least one 4KB block allocated.

So, it is _possible_ to reduce the inode *ratio* below 2.5KB if you know what you are doing (e.g. 2KB/inode or 1.5KB/inode, this can be an arbitrary number of bytes, it doesn't have to be an even multiple of anything) but it definitely isn't possible to have 1KB inode size and 1KB per inode ratio, as there wouldn't be *any* space left for directories, log files, journal, etc.

Cheers, Andreas
--
Andreas Dilger
Principal Lustre Architect
Whamcloud





-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lustre.org/pipermail/lustre-discuss-lustre.org/attachments/20191003/3d5ddb68/attachment-0001.html>


More information about the lustre-discuss mailing list