toasters February 2013

toasters@lists.teaparty.net

44 participants
30 discussions

Mixed CIFS/NFS Access and issues w/ Locking
by Ray Van Dolson 21 Feb '13

21 Feb '13

We have an ONTAP 8.0.x filer here serving out a volume with an NTFS qtree via both CIFS and NFS. We're encountering the following situation: 1. An OSX user browses to the path via CIFS in Finder and then walks away/moves on to another task leaving the Finder window open on the path. - This generates a mandatory read lock on the folder (verified via lock status): CIFS path=\path(/path) host=10.28.102.30() owner=njarecha state=GRANTED mode=Read-denyN oplock=None 2. An NFS client comes along now (automated build process) and attempts delete the same folder. This appears to succeed from the NFS client's perspective. - We now see the following: CIFS path=\path(/path) host=10.28.102.30() owner=njarecha state=GRANTED mode=Read-denyN oplock=None CIFS path=\path(-) host= owner= state=GWAITING mode=DelOnClose At this point, CIFS access to the path now is completely blocked. No one can navigate into the folder, nor view its ACLs (save likely for the OSX client still holding the read lock). NFS clients can continue to access the contents of the folder (both read/write), but rmdir/rm operations on the folder itself fail with Permission denied. The fix is to (a) close the Finder window or (b) clear the lock via 'lock break' on the filer. I'll note that Windows Explorer doesn't appear to open a read lock when you only browse to a path -- not sure why Finder does. We're looking for options here, but first I'm wondering if the filer's response here isn't ideal. Rather than completing closing off access to CIFS clients, shouldn't #2 above just fail from the get-go rather than generating the DelOnClose queued event (if a mandatory read lock exists on the resource)? Other options we're considering: - Modifying automated processes to avoid the "rmdir" command. - Sticking with only one access protocol (likely CIFS). - Modifying OSX Finder behavior to not generate and hold a CIFS read lock (not sure how easy this would be to do). - Finding some sort of tunable on the ONTAP side to minimize impact (e.g. ignore the read lock and let the NFS rmdir operation to succeed) at the potential elevated risk of reduced data integrity. Anyone have any thoughts/experiences with something similar? Thanks, Ray

3 4

can't shoot SSH commands to a filer
by Randy Rue 20 Feb '13

20 Feb '13

Hello All, I have a 3020 running 7.2.4 that is failing to respond to SSH commands when they're sent one at a time. I can ssh in as root and get a shell via putty or ssh. But if I try to send an "uptime" command I get either "FATAL ERROR: Server unexpectedly closed network connection" (from plink) or "Connection to carbon-b closed by remote host." from ssh. Also, when I'm logged in interactively, I see lots of this blowing across the console: carbon-b> receive signals: rsh protocol failure in circuit setup. 300 received receive signals: rsh protocol failure in circuit setup. 300 received Try again. Try again. receive signals: rsh protocol failure in circuit setup. 300 received receive signals: rsh protocol failure in circuit setup. 300 received receive signals: rsh protocol failure in circuit setup. 300 received receive signals: rsh protocol failure in circuit setup. 300 received receive signals: rsh protocol failure in circuit setup. 300 received Try again. receive signals: rsh protocol failure in circuit setup. 300 received Try again. receive signals: rsh protocol failure in circuit setup. 300 received Try again. receive signals: rsh protocol failure in circuit setup. 300 received Try again. Try again. Try again. receive signals: rsh protocol failure in circuit setup. 300 received Try again. Try again. Try again. Try again. Try again. Try again. Try again. Try again. receive signals: rsh protocol failure in circuit setup. 300 received Try again. receive signals: rsh protocol failure in circuit setup. 300 received Try again. receive signals: rsh protocol failure in circuit setup. 300 received Try again. Try again. receive signals: rsh protocol failure in circuit setup. 300 received Try again. Try again. Try again. Try again. Try again. Try again. receive signals: rsh protocol failure in circuit setup. 300 received Try again. Try again. Try again. Try again. Try again. receive signals: rsh protocol failure in circuit setup. 300 received Try again. Try again. Try again. Try again. Try again. Try again. receive signals: rsh protocol failure in circuit setup. 300 received We have lots of processes (monitoring, alerting, scripted processes) that touch this box via non-interactive SSH, is each of these errors some failed attempt to connect? If I understand the inner workings correctly, pretty much all admin traffic on a filer channels into an internal RSH pipe. Are there different pipes for interactive and non-interactive SSH? I have tried stopping and restarting SSH (options ssh.enable off and on). Also RSH (options rsh.enable off and on). No love. Anyone have any ideas? Randy Seattle

2 2

Migration of Oncommand Host services
by KICKEN Ben (INT/TEI) 15 Feb '13

15 Feb '13

Greetings toasters, I am looking for migration procedures for the Oncommand Host Services, installed on the DFM/OCUM server (which is a VM) to the vCenter server (also a VM), without losing all configured datasets, protection and provisioning policies which Snapvault primary data to secondary storage. I was thinking of installing Host Package 1.2 on vCenter server and authorize it on the OCUM host - which will present all Vmware objects twice in the OC COnsole. Next unregister the OCUM hostservice (dfm hs unregister). Will all abovementioned datasets and policies still be available and active, on the hostservice running on the vCenter Host? Unfortunately I have no test environment for simulating this before doing this in real life; asked this in Communities and created a case at netapp support, but no good suggestions yet. thanks for any input Ben

2 1

Live disk shelf removal.
by Jeff Cleverley 13 Feb '13

13 Feb '13

Greetings, Is there a way to remove DS14 MK4 shelves from a running cluster? I need to look into isolating some drives and removing some shelves and would like to do it without having to halt the cluster. This is a 7.3.5.1 Ontap cluster on FAS6080s. Thanks, Jeff -- Jeff Cleverley Unix Systems Administrator 4380 Ziegler Road Fort Collins, Colorado 80525 970-288-4611

6 6

HTTP & CFO
by Ray Van Dolson 13 Feb '13

13 Feb '13

ONTAP 8.0.x with httpd.enable set and virtual hosting configured via httpd.hostprefixes. Our PRD filer in question has two controllers in CFO mode which works fine for NFS/CIFS services. Will the HTTP services fail over properly as well? Tried to test this with the ONTAP 8.0.4 simulator, but you can't test CFO there apparently. Am especially curious about the httpd.hostprefixes settings and if they'll fail over gracefully. Thanks, Ray

1 0

can't touch a CIFS folder?
by Randy Rue 12 Feb '13

12 Feb '13

Hello All, Got a subfolder on a CIFS only vfiler (NTFS qtree) that I can't open, delete or rename, even though I'm connected as a user that's a member of the local administrators for the vfiler. From a windows host if I right-click the folder and choose properties there isn't even a permissions tab. Can't seize ownership. Can't open the folder (denied). >From a cifs-mounted linux host I also strike out but with odd errors. [root@host]# mv _DIVA\ templates delete mv: cannot stat `_DIVA templates': No such file or directory [root@host]# ls _DIVA templates [root@host]# ls -al total 24 drwxrwxrwx 3 root root 0 Feb 11 21:07 . drwxrwxrwx 28 root root 0 Jan 28 12:23 .. ?--------- ? ? ? ? ? _DIVA templates [root@host]# Note that the folder is called "_DIVA templates" I've tried copying and pasting it a few different ways to see if maybe it had funky/non-printable characters in the name but it doesn't appear so. Any ideas? Randy Rue Seattle WA

5 6

6080 heads with 6040 NVRAM cards.
by Jeff Cleverley 12 Feb '13

12 Feb '13

Greetings, I'm thinking about doing something that is not supported and was wondering if anyone had done the same or has more detailed insight. We have a very busy cluster (6040s 7.3.5.1P4). It looks like we are largely maxing out the heads for CPU. We are getting a pair of 6080s and really need to try and do the head swap live (takeover / giveback) if at all possible. The unsupported part I want to do is keep the 6040 NVRAM cards and put them in the 6080s as I swap them. The reason for this is I would not have to change the system ID ownership on all the drives. I know changing the system ID is generally not a big deal by booting each head to maintenance mode and reassigning the old SID to the new SID. In our case it worries me. Last week we were going to move a project to the other head by reassigning the appropriate drives for a couple of aggregates. While trying to reassign these the SAS buses started panic'ing and crashed the controlling filer. The entire cluster was down. The ensuing mess took several hours to clean up. If it crashed while trying to change ownership of a few drives, I'm afraid of what will happen when it tries to reassign all the old SID drives for the new NVRAM card. I was hoping if we could keep the cards, we could swap heads, not change SIDs, and minimize our chance of repeating the crash. I could do the disks one at a time, but I have 796 drives on this cluster and would rather not. Is there a requirement for the hardware to have the bigger memory cards? Since there are more CPUs, I can see where maybe something needs it, I just don't know what. We will probably have a downtime in a couple of months where I can put the correct ones back in. Thanks, Jeff -- Jeff Cleverley Unix Systems Administrator 4380 Ziegler Road Fort Collins, Colorado 80525 970-288-4611

5 6

OnTap 8 Cluster install
by René Bormann 11 Feb '13

11 Feb '13

Hello Toasters, in our lab I got a FAS3070 Cluster to play around with. So this friday afternoon I wanted to install OnTap 8.1.2 Cluster over my 7-Mode test-installtion. Data loss is no problem. I tried to install the cluster software over the 7-mode Version with software update several times, but on every reboot. Ontap greets me with "7-Mode". Which point am I missing? Thanks, René

6 5

Merging Snapvault Volumes
by Steven Kreuzer 08 Feb '13

08 Feb '13

I opened up a case with NetApp about this, but I figured I should also post here and see what the community has to say. In a nutshell, I have two separate clusters and a volume with home directories on Filer_A that I was snapvaulting to Filer_B. During the hurricane, I had to shut down Filer_A, break the snapvault relationship on Filer_B to promote the volume to RW and then started to have everyone use Filer_B as their home. Folks have been working on Filer_B for 3 or so months now but I am ready to flip back and make Filer_A the primary once again. What are my options for basically taking all the new data as well as the snapshots that are on Filer_B and syncing them back to Filer_A, without losing any of the snapshots that currently exist on Filer_A? Right now the only option that comes to mind is the use rsync and copy a snapshot from Filer_B to Filer_A, take a snapshot on Filer_A, and repeat that process over and over until they have all been recreated on Filer_A. With that, does anyone have any better suggestions? Thanks in advance

5 6

Nifty ONTAP option nfs.max_num_aux_groups
by Steve Losen 08 Feb '13

08 Feb '13

Hi folks, We tend to use unix style security on shares that are accessed via both NFS and CIFS. Being a university we have a large number of users and groups, and some folks belong to numerous groups. (We use group permissions for shared projects and some folks are members of many unix groups.) We use NFS v3 exclusively and the auxiliary group list is passed by the NFS client in each NFS packet, and it is limited to 16 groups. This is rather inconvenient for some NFS users. Does anyone know if this is a hard NFS v3 limit? For CIFS we authenticate via a Windows domain and the CIFS credentials are mapped to Unix credentials for unix style security. We noticed with "cifs shares -s username" that folks were being limited to 32 unix groups. I have discovered that you can do this: options nfs.max_num_aux_groups 256 (the only legal values are 32 and 256, we have DOT 8.0.1 7-Mode.) and now our CIFS users who belong to over 32 unix groups are getting all their groups. No help for NFS v3 but this will make several of our CIFS users happy. Steve Losen scl(a)virginia.edu phone: 434-924-0640 University of Virginia ITC Unix Support

3 2

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

toasters February 2013