avati, In prior versions attempting to run glusterfs without first running umount resulted in an error. Now it double mounts. I am not sure in which patch this behavior changed but I do know that there was an error message because sometimes I forget to umount before restarting glusterfs. Harris ----- Original Message ----- From: "Anand Avati" <avati@xxxxxxxxxxxxx> To: "Harris Landgarten" <harrisl@xxxxxxxxxxxxx> Cc: "gluster-devel" <gluster-devel@xxxxxxxxxx> Sent: Wednesday, July 4, 2007 11:52:26 AM (GMT-0500) America/New_York Subject: Re: problems with patch 268 Harris, the size calculation was a bug introduced in patch 268 with optimum block size tuning. however the double mount clearly seems to be an issue with your setup. most likely you did an overmount over the old mountpoint! please double check. patch-269 fixes du output. (sorry your storage just came down by ~73.7 TB :p) avati 2007/7/4 , Harris Landgarten < harrisl@xxxxxxxxxxxxx >: df -h output: With patch 267: glusterfs 294G 20G 259G 8% /mnt/glusterfs With patch 268: glusterfs 74T 5.0T 65T 8% /mnt/glusterfs Harris ----- Original Message ----- From: "Harris Landgarten" < harrisl@xxxxxxxxxxxxx > To: "gluster-devel" < gluster-devel@xxxxxxxxxx > Sent: Wednesday, July 4, 2007 11:18:16 AM (GMT-0500) America/New_York Subject: problems with patch 268 Patch 268 introduces the following new problems: 1. on client glusterfs runs with /mnt/glusterfs already mounted with prior version. This results in the mount point mounted twice. 2. after umounting twice and running glusterfs again it reports that a 294G unify volume has 5T. 3. reverting to patch 267 corrected the problem Harris _______________________________________________ Gluster-devel mailing list Gluster-devel@xxxxxxxxxx http://lists.nongnu.org/mailman/listinfo/gluster-devel _______________________________________________ Gluster-devel mailing list Gluster-devel@xxxxxxxxxx http://lists.nongnu.org/mailman/listinfo/gluster-devel -- Anand V. Avati
