[lustre-discuss] lustre 2.5.3 changelog mask question

Kurt Strosahl strosahl at jlab.org
Fri May 29 08:06:04 PDT 2015


OK,

   Good to know.  But someone over at Intel needs to update that documentation.

w/r,
Kurt

----- Original Message -----
From: "Olaf P. Faaland" <faaland1 at llnl.gov>
To: "Kurt Strosahl" <strosahl at jlab.org>, lustre-discuss at lists.lustre.org
Sent: Friday, May 29, 2015 11:04:17 AM
Subject: Re: [lustre-discuss] lustre 2.5.3 changelog mask question

Kurt,

The 2.5.3 code appears to me to support the following list of changelog
mask values:

        static const char *changelog_str[] = {
                "MARK",  "CREAT", "MKDIR", "HLINK", "SLINK", "MKNOD",
"UNLNK",
                "RMDIR", "RENME", "RNMTO", "OPEN",  "CLOSE", "LYOUT",
"TRUNC",
                "SATTR", "XATTR", "HSM",   "MTIME", "CTIME", "ATIME",
};


IOCTL went away between 2.3 and 2.4, see LU-3279
RNMFM went away between 2.2.90 and 2.2.91, see LU-1331

-Olaf

On 5/29/15, 6:45 AM, "Kurt Strosahl" <strosahl at jlab.org> wrote:

>Good Morning,
>
>   I'm trying to set the changelog mask on a lustre 2.5.3 system and it
>seems that the documentation isn't 100% accurate.
>
>According to the documentation found at
>https://build.hpdd.intel.com/job/lustre-manual/lastSuccessfulBuild/artifac
>t/lustre_manual.pdf one of the values is IOCTL.  However when I try to
>set the mask I get the following error...
>
>lctl set_param mdd.lustre2-MDT0000.changelog_mask="MKDIR CREAT HLINK
>UNLNK RMDIR RENME RNMTO TRUNC SATTR IOCTL"
>mdd.lustre2-MDT0000.changelog_mask=MKDIR CREAT HLINK UNLNK RMDIR RENME
>RNMTO TRUNC SATTR IOCTL
>error: set_param: setting
>/proc/fs/lustre/mdd/lustre2-MDT0000/changelog_mask=MKDIR CREAT HLINK
>UNLNK RMDIR RENME RNMTO TRUNC SATTR IOCTL: Invalid argument
>
>The same seems to be true for RNMFM.
>
>What was that value changed to?
>
>w/r,
>Kurt
>_______________________________________________
>lustre-discuss mailing list
>lustre-discuss at lists.lustre.org
>http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org


More information about the lustre-discuss mailing list