| To: | Chris Wedgwood <cw@xxxxxxxx> |
|---|---|
| Subject: | Re: validating root filesystem ... |
| From: | "L. A. Walsh" <xfs@xxxxxxxxx> |
| Date: | Sat, 04 Feb 2006 17:53:28 -0800 |
| Cc: | Linux-Xfs <linux-xfs@xxxxxxxxxxx> |
| In-reply-to: | <20060203065747.GA17620@xxxxxxxxxxxxxxxxxxxxx> |
| References: | <43E1C5C6.9080005@xxxxxxxxx> <20060203065747.GA17620@xxxxxxxxxxxxxxxxxxxxx> |
| Sender: | linux-xfs-bounce@xxxxxxxxxxx |
| User-agent: | Thunderbird 1.5 (Windows/20051201) |
Chris Wedgwood wrote: On Thu, Feb 02, 2006 at 12:41:42AM -0800, L. A. Walsh wrote:Anyone know why I would get a persistent message out of xfs_check?Don't run it. Especially not on a mounted filesystem, even mounted RO. --- Hi Chris...then, this begs the question:How should one check or repair a root file system? I believe I understand that mounting a root file system "ro" will still replay any outstanding log operations. However, after those operations are played, if we only have "r/o" specified in the mount options, then future actions on the disk should be limited to "read" operations, no? Isn't the journal only used when "writing" to disk"? I'm not clear why xfs_check or xfs_repair shouldn't be usable to validate the root disk image. Thanks, Linda |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: xfs_check: sb_fdblocks x, counted x+1, Chris Wedgwood |
|---|---|
| Next by Date: | TAKE 904196 - Merge up to 2.6.16-rc2., Nathan Scott |
| Previous by Thread: | Re: xfs_check: sb_fdblocks x, counted x+1, Chris Wedgwood |
| Next by Thread: | Re: validating root filesystem ..., Chris Wedgwood |
| Indexes: | [Date] [Thread] [Top] [All Lists] |