[SERVER-27221] Startup warning about using XFS doesn't actually check for XFS Created: 30/Nov/16 Updated: 06/Dec/22 Resolved: 24/Jun/19 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Storage |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Jonathan Reams | Assignee: | Backlog - Storage Execution Team |
| Resolution: | Won't Do | Votes: | 1 |
| Labels: | neweng | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||
| Assigned Teams: |
Storage Execution
|
||||||||
| Operating System: | ALL | ||||||||
| Participants: | |||||||||
| Description |
|
The following warning introduced by
We should change this to check that fs_stats.f_type != XFS_SUPER_MAGIC instead of fs_stats.f_type == EXT4_SUPER_MAGIC. |
| Comments |
| Comment by Chad Kreimendahl [ 03/Dec/16 ] | |||
|
It's not the warning message. It's the if statement. The original bug and the other comments are absolutely clear on this point. Do you not recommend people use xfs over FAT or btrfs? If (NOT XFS) is the logical setup. the problem is you need to ask if it's not xfs (as others have stated), instead of only complaining about ext. More specifically I can assure you btrfs and fat are even worse for performance than ext4. We tested extensively. | |||
| Comment by Eric Milkie [ 03/Dec/16 ] | |||
|
The "if" statement is not illogical at all. If the file system for the dbpath is ext2, ext3, or ext4, the code displays a warning saying that xfs is recommended instead. chad@onspring.com sallgeud are you suggesting that we should recommend a different filesystem other than xfs in the warning message? I have already stated that users running other filesystems than ext2, ext3, or ext4, should not receive a warning. Personally, I believe the warning message is fine, as it clearly communicates what the administrator needs to do to clear the warning (but it could be made more explicit with Kevin's suggested wording). I don't see how the warning or the code has no purpose. | |||
| Comment by Chad Kreimendahl [ 02/Dec/16 ] | |||
|
Yes. The if statement is illogical relative to the message it generates. It's clearly a bug. While insignificant to functionality, it is almost entirely without purpose the way it's written. | |||
| Comment by Jonathan Reams [ 02/Dec/16 ] | |||
|
My point is that the error message is not connected to the check that triggers to it. We're telling you to use the XFS filesystem, but we check that you're using EXT4. | |||
| Comment by Eric Milkie [ 02/Dec/16 ] | |||
|
I'm in favor of changing the message, but I don't understand the incentive to change the logic. If a user is electing to use a filesystem other than EXT* or XFS on Linux, they already have to be expecting that they are going to have a performance profile that might not be ideal. | |||
| Comment by Kevin Pulo [ 30/Nov/16 ] | |||
|
I also recommend we adjust the message a little to make it clearer what the actual problem is, eg:
|