[lustre-discuss] Lustre Orphaned Chunks
Hi All. I am still learning Lustre and I have run into an issue. I have referred to both the Lustre admin manual and Google, but I've had no luck in finding the answer. We are using Lustre 2.8.0. We had an OST fill due to a single large file. I took the OST offline via the lctl deactivate command to prevent new files from being created on the OST. While the OST was deactivated, the user deleted the file. Now it appears the metadata is gone from the MDS (which makes sense), but the data chunks on the OST remain even after I reactivated the OST. I believe that running lfsck would resolve this issue, but I am not sure if I should run it on the MDS or the OST? If this is the fix, what options would I need to use? Thank you in advance, Chad Chad DeWitt, CISSP | HPC Storage Administrator UNC Charlotte *| *ITS – University Research Computing ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
Re: [lustre-discuss] Lustre Orphaned Chunks
Hi All, I wanted to follow up and explain how I solved the issue in case anyone else encounters this situation. All environments are different, so YMMV. (Please note per my initial email, the problematic OST is marked as active before performing these steps.) First, I marked the problematic OST as degraded (on the OSS): # lctl set_param obdfilter..degraded=1 Second, I kicked off a lfsck for the Lustre filesystem (on the MDS): # lctl lfsck_start --orphan --device Once the lfsck had deleted the orphaned chunks, I marked the OST as normal (on the OSS): # lctl set_param obdfilter..degraded=0 In /var/log/messages on the OSS, I could see the orphans were deleted: kernel: Lustre: : deleting orphan objects from 0x0:162957594 to 0x0:162957873 I would like to thank Shawn [Hall] and Bob [Ball] for their responses, which lead me in the right direction. Thank you, Chad Chad DeWitt, CISSP | HPC Storage Administrator UNC Charlotte *| *ITS – University Research Computing On Mon, Oct 17, 2016 at 2:32 PM, DeWitt, Chad wrote: > Hi All. > > I am still learning Lustre and I have run into an issue. I have referred > to both the Lustre admin manual and Google, but I've had no luck in finding > the answer. We are using Lustre 2.8.0. > > We had an OST fill due to a single large file. I took the OST offline via > the lctl deactivate command to prevent new files from being created on the > OST. While the OST was deactivated, the user deleted the file. Now it > appears the metadata is gone from the MDS (which makes sense), but the data > chunks on the OST remain even after I reactivated the OST. > > I believe that running lfsck would resolve this issue, but I am not sure > if I should run it on the MDS or the OST? If this is the fix, what options > would I need to use? > > Thank you in advance, > Chad > > > > > Chad DeWitt, CISSP | HPC Storage Administrator > > UNC Charlotte *| *ITS – University Research Computing > > > ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
[lustre-discuss] Lustre [2.8.0] flock Functionality
Good afternoon, All. We've encountered several programs that require flock, so we are now investigating enabling flock functionality. However, the Lustre manual includes a passage in regards to flocks which gives us pause: "Warning This mode affects the performance of the file being flocked and may affect stability, depending on the Lustre version used. Consider using a newer Lustre version which is more stable. If the consistent mode is enabled and no applications are using flock, then it has no effect." We are running Lustre 2.8.0 (servers and clients). I've looked through Jira, but didn't see anything that looked like a showstopper. Just curious if anyone has enabled flocks and encountered issues? Anything in particular to look out for? Thank you in advance, Chad Chad DeWitt, CISSP | HPC Storage Administrator UNC Charlotte *| *ITS – University Research Computing ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
