It has been integrated into mbralign. Try "mbralign --help" -Michael ... View more

You have to do it client-side, there's no way around that. There, it depends on what systems you have connected. ESX is easy, you can do it with Storage VMotion. Otherwise you probably need to create new LUNs and copy all files over somehow I'm not aware of any product that lets you migrate from 3rd party storage to NetApp -Michael ... View more

I can only answer part of your question: Each filer has its own spare disks. So it's perfectly normal to see different spare counts for each filer. However, why the partner filer still thinkgs that 1c.58 is FAILED although the owning filer (filerB) sees the disk as working is beyond me. Maybe a simple "disk fail" followed by "disk unfail" (in diag mode) on filerB can resolve that issue. Otherwise I would open a case with NetApp if it bothers you too much. But as long as the owning filer sees the disk as usable (and not the other way round) it's normally not a problem -Michael ... View more

No, it means you cannot hot-add the shelf to a loop containing FC shelves. You can, of course, hot-add a new DS14 shelf to a completely new FC port on your filer, as long as tthis port is configured as INITIATOR (not TARGET). See, for example, the DS14mk2 AT Hardware Service Guide (https://now.netapp.com/NOW/knowledge/docs/hardware/filer/210-02360_F0.pdf) on page 48, chapter "Hot-adding a disk shelf to an existing adapter in your system" hope that helps -Michael ... View more

You can use the Space Reclaimer from within SnapDrive to accomplish the same objective. And with LUNs you still have the same problem that any free space that you get in the volume (regardless of what triggers it: TRIM, Deduplication, ...) is not immediately available to ESX and you have to use LUN overcommitment to effectively use that free space. BTW I think it will take a bit more than "a few days" for vSphere 5 to be released. Last time I heard they were talking about a March 2010 release or, somwhere around March -Michael ... View more

What do you mean by "firmware for VMware"? Firmware is for hardware, so you probably mean BIOS, which you'll have to check with your server vendor. Or maybe you mean firmware for an FC HBA or iSCSI HBA, in which case you'd have to check the VMware HCL and/or look on the HBA's vendor pages If you can be more specific on what you're looking for, maybe we can help you better -Michael ... View more

This happens if you delete files (or snapshots) from inside your LUN. ESX only marks the blocks as "free" but on the NetApp, they still contain the old data and are thus "used". There's no easy way to "reclaim" this space on the filer, at least not on ESX. -Michael ... View more

eric.barlier wrote: With respect to your problem. As mentioned you cannot have 1 disk on one controller. the least amount you can have is 2 for a raid4 aggr. to accomodate a root volume. Just some nitpicking, it is perfectly possible from a technical point of view to have a single-disk root aggregate. However this is not documented, not supported and *will* get you into trouble if you try and open up a support case 😉 netapp10> sysconfig -r Aggregate aggr0 (online, raid0) (block checksums)   Plex /aggr0/plex0 (online, normal, active, pool0)     RAID group /aggr0/plex0/rg0 (normal)       RAID Disk Device  HA  SHELF BAY CHAN Pool Type  RPM  Used (MB/blks)    Phys (MB/blks)       --------- ------  ------------- ---- ---- ---- ----- --------------    --------------       data      0c.00.2 0c    0   2   SA:B   0  SATA  7200 423111/866531584  423946/868242816 Pool1 spare disks (empty) Pool0 spare disks (empty) Partner disks RAID Disk       Device  HA  SHELF BAY CHAN Pool Type  RPM  Used (MB/blks)    Phys (MB/blks) ---------       ------  ------------- ---- ---- ---- ----- --------------    -------------- partner         0c.00.3 0c    0   3   SA:B   0  SATA  7200 0/0               423946/868242816 partner         0c.00.6 0c    0   6   SA:B   0  SATA  7200 0/0               423946/868242816 partner         0c.00.0 0c    0   0   SA:B   0  SATA  7200 0/0               423946/868242816 partner         0c.00.1 0c    0   1   SA:B   0  SATA  7200 0/0               423946/868242816 partner         0c.00.7 0c    0   7   SA:B   0  SATA  7200 0/0               423946/868242816 partner         0c.00.5 0c    0   5   SA:B   0  SATA  7200 0/0               423946/868242816 partner         0c.00.4 0c    0   4   SA:B   0  SATA  7200 0/0               423946/868242816 ... View more

since this is a per-share setting you can change it for every CIFS share that you have. From the CLI you can do it with cifs shares -change SHARENAME -vscan -novscanread This command changes the policy to only scan on writes, not on reads. There's more info about CIFS virus scanning in the "Data Protection Online Backup and Recovery Guide" -Michael ... View more

I don't know any tool for viewing the most active files from the top of my head. but there is "nfsstat -d" which shows you misaligned vmdks, and these are more often than not the source of increasing performance problems (misaligned IOs multiply the IOs that your disks have to do) On the ESX side there's ESXtop which can produce files that can be opened and viewed in windows perfmon. There you see the IO generated by each vm, or on each HBA/Datastore hope that helps -Michael ... View more

If you're in diag mode ("priv set diag") you get many more counters to monitor. Included are lots of SIS counters: fas3020c2*> stats list counters sis Counters for object name: sis         state         status         progress         schedule         last_operation_begin         last_operation_end         last_operation_size         last_operation_error         changelog_usage         stage         num_fingerprint_gathered         gathering_begin         num_fingerprint_sorted         num_duplicate_blks_found         sorting_begin         num_blks_deduplicated         dedup_begin         num_fingerprint_deleted         checking_begin         saving         percent_saved         num_succeeded_op         num_started_op         num_deferred_op         num_stopped_op         num_failed_op         total_sorted_blocks         num_same_fingerprint         num_max_reference_hits         num_staled_recipient         num_staled_donor         num_file_too_small         num_out_of_space         total_blk_cnt         total_op_time         total_verify_time         num_fp_deleted         blks_used         blks_shared         share_saved_blks         num_sis_files         num_blks_mismatch         num_deleted_inode_records         num_blks_vol_dense_metadata         ref_count_histogram         block_sharing_histogram         total_blocks_processed_history         total_time_taken_history         num_stale_blks_mismatch fas3020c2*> stats list instances sis Instances for object name: sis         /vol/slesxX_ATA         /vol/slesxX_LocalEnvCVLT         /vol/slesxX_LocalEnvEMA         /vol/slesxX_LocalEnvNAE3S2         /vol/slesxX_LocalEnvNAE5         /vol/slesxX_LocalEnvNAE6 fas3020c2*> -Michael ... View more

could you try it with only the secondary path? as in filer> snapvault release /vol/ossv_backup01/obsljtdb02 also, maybe post the output of "snapvault status" and "snapvault destinations", too. I don't think it's a bug, this would be a rather big one and thus it would be noticed almost instantly -Michael ... View more

I think there's something wrong with what you pasted here. It says that your primary path is "owbsljtdb02:/" and your secondary path is "dcbsnap02:/vol/ossv_backup01/obsljtdb02". But snapvault release tells you that the first argument should be the secondary path and the second argument the primary. So I *think* you have to type   snapvault release /vol/ossv_backup01/obsljtdb02 owbsljtdb02:/ on your filer (sorry, but I don't have an OSSV system here to test at the moment) -Michael ... View more

Your primary path is something like FILER:/vol/volumeXY/qtreeZ, or in the case of OSSV, something like SERVER:/C:/ If in doubt, just copy/paste the output of "snapvault status" -Michael ... View more

Again, you're mixing things up. The 100k entries per directory limit as nothing to do with hard vs. soft links. It simply means you cannot have more than 100000 entries in one directory (minus the 2 entries that are always there for . and .., so in reality you can only have 99998 files). It really doesn't matter if it's hardlinks, softlinks, subdirectories, or anything else. The limit is 99998 *entries* in the WAFL directory file. The other limit that was mentioned above (Bug 292410) is something different, this one talks about hardlinks to one file (which can reside in different directories), and there, too, is a limit of 100000 hardlinks. -Michael ... View more

From my experience, the problem is worse with >90% full volumes than it is with >90% full aggregates. If the volume is too full, the background reallocation task can't do it's job correctly, and manual defragmentation (with "reallocate start") will also be impossible. This is the only reason why I always suggest keeping the volumes below 85-90%. Reallocation runs in each volume and needs free space to work with -Michael ... View more

Basically, you want to try to avoid that both the disks and the controller stop a the same time.  This makes it easier (possible) for the OS to make decisions concerning a fail-over.  When everything goes at the same time, you get a "split-brain" situation.  Just some nit-picking: No, you won't get into a split-brain situation. You never get into that with NetApp. You get in a situation where the filer cannot be 100% sure that a takeover will NOT result in a split brain. "split brain" means you have two copies of your data active, one in each site. This *could* happen if the controller takes over automatically, but since it won't (you have to do "cf forcetakeover -d") you don't get into the split brain situation. -Michael ... View more

Check if fcp is running at all ("fcp status"). Also check to see what initiators your filer sees as connected ("fcp show initiators") Finally check that you have configured your (onboard-)ports as target ports ("fcadmin config") -Michael ... View more

Depends on the throughput to your datastore. If you use VMFS (LUN) datastores, I wouldn't reconmmend aligning more than 1 VM per ESX server (as the bandwidth from the service console is very limited). If you have NFS datastores and are aligning from a Linux system (i.e. not the service console) you can probably align 4-8 or more per filer, depending on bandwidth to your filer. -Michael ... View more

VSM is not possible between 32 and 64 bit. Even if you *could* move your destination from 32 to 64 bit, your SnapMirror Relationship would not work anymore because your source is still 32bit. You need to migrate both the source and the destination to 64bit, and then it is probably faster to just migrate the source and do a new baseline transfer (via LREP/SM2T if your bandwidth is limited) -Michael ... View more