[LU-2814] discover the server version from the client Created: 14/Feb/13 Updated: 07/Jun/17 Resolved: 30/May/17 |
|
| Status: | Resolved |
| Project: | Lustre |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | Lustre 2.6.0 |
| Type: | Bug | Priority: | Minor |
| Reporter: | Richard Henwood (Inactive) | Assignee: | WC Triage |
| Resolution: | Fixed | Votes: | 0 |
| Labels: | hackathon2017 | ||
| Issue Links: |
|
||||||||
| Severity: | 3 | ||||||||
| Rank (Obsolete): | 6818 | ||||||||
| Description |
|
You can discover the client software version using: # lctl get_param version lustre: 2.3.61 kernel: patchless_client build: v2_3_61_0-g997d2b5-CHANGED-2.6.32 However, the version of the server software is not readily available to a user without access to the servers. The client knowns what version of the servers it is connected to. It would be good to expose the server version on the client. A starting point for this information would be struct obd_connect_data:ocd_version |
| Comments |
| Comment by Andreas Dilger [ 14/Feb/13 ] |
|
One option is to have /proc/fs/lustre/ {osc,mdc}/*/import also print out the data fields in obd_connect_data, in addition to just the flags. This would be useful for a number of reasons, in addition to just getting the server versions. I was just thinking that it would be good for getting the ocd_brw_size during 4MB patch testing, etc. Something like: import:
name: myth-OST0000-osc-ffff88004fc2d800
target: myth-OST0000_UUID
state: FULL
connect_flags: [write_grant, server_lock, version, request_portal, truncate_lock, max_byte_per_rpc, early_lock_cancel, adaptive_timeouts, lru_resize, alt_checksum_algorithm, version_recovery, full20, 64bithash, object_max_bytes]
connect_data:
version: 2.1.3
grant: 2097152
index: 0
brw_size: 4194304
ibits_known: 0xf
blocksize: 4096
inodespace: 0
grant_extent: 24
transno: 0
group: 0
cksum_types: [ crc32, adler, crc32c ]
max_easize: 4048
instance: 4
maxbytes: 17179869184
Note that the "instance:" field should be moved into the "connect_data" section rather than being a separate field. |
| Comment by Robert Read (Inactive) [ 14/Feb/13 ] |
|
Was just about to suggest the same thing. |
| Comment by Andreas Dilger [ 30/May/17 ] |
|
This was fixed with patch http://review.whamcloud.com/6588 " |