Re: [lustre-discuss] FID error showing on MDS

Please, Could someone give a short answer about the need to worry or if it is harmless ? Thank you all Le 21/09/2016 10:09, Jérôme BECOT a écrit : Hello, We migrated server software from 2.6 to 2.7 last month and now logs are filled with these lines : lustre-MDT-osd: FID [0x25ed

Re: [lustre-discuss] refresh file layout error

inline On 9/26/16, 10:34 AM, "lustre-discuss on behalf of E.S. Rosenberg" mailto:lustre-discuss-boun...@lists.lustre.org> on behalf of esr+lus...@mail.hebrew.edu> wrote: Which leads me to 2 questions about current development: 1. Has the mainline kernel seen

Re: [lustre-discuss] refresh file layout error

I hope no-one minds me restoring this thread, I just wanted to report back: After we switched from vanilla kernel-staging lustre to building from source (lustre 2.8.0 + kernel 4.2.8) this problem ceased to exist, so it must have been an issue of lustre 2.3.64 (kernel) clients not talking properly

Re: [lustre-discuss] Getting Monthly file usage report on lustre 1.8.7

'lfs df' and 'lfs df -I' will give you OST-level information on bytes used and files used. Those are both pretty rough numbers, though. Other than that, I believe you need to use normal POSIX tools for scanning directories and gathering information. Depending on the size of your system, that

Re: [lustre-discuss] lfs_migrate transfer question

Pre-make your list of files, split it into 5 or so parts, give each of 5 clients one piece, and let them migrate in parallel. bob On 9/26/2016 9:00 AM, Jérôme BECOT wrote: Hello, As we had a inode usage issue, we did as Andreas advised : - add a new ost with much more inodes - disable the fi

[lustre-discuss] lfs_migrate transfer question

Hello, As we had a inode usage issue, we did as Andreas advised : - add a new ost with much more inodes - disable the first ost - use lfs_migrate to move files to the new ost We have about 16M files on this OST (7.5TB). Our OSS are connected via 10Gbps Ethernet, though the lfs_migrate command