toasters

toasters@lists.teaparty.net

3 participants
13514 discussions

Re: What do you monitor
by Nick Bernstein 03 Oct '10

03 Oct '10

I was mainly looking for things that weren't just the obvious capacity planning / performance stuff, but this has been a good list - thanks everyone. I took people's responses and dumped them into a wiki, if people are interested in referring back to this later on. Feel free to add/edit. http://nicholasbernstein.com/netapp/wiki/doku.php?id=monitoring Cheers! Nick On Fri, Oct 1, 2010 at 5:59 AM, Collins, Brian <brianc(a)netapp.com> wrote: > Funny, nobody has mentioned volume capacity..... Just sayin' > > > On 9/30/10 11:38 AM, "Nicholas Bernstein" <nick(a)nicholasbernstein.com> > wrote: > >> I had an interesting question posed by a student yesterday - "aside >> from interfaces being up, and volumes being online, what do you >> typically monitor?" Some of my initial thoughts were: >> >> - inodes being available >> - ifstat on multimode vifs to make sure interfaces are being used >> - io counts on luns/qtrees/vols/ >> - client side nfsstat or the equivalent for that protocol from the >> client side >> >> Anyway, I was trying to think of some of the "non-standard" things to >> monitor, and thought I'd put it out to the list and see what other >> people are typically doing. >> >> Cheers, >> Nick > > -- > Brian M Collins > Virtualization Solutions Architect > NetApp, Inc. > +1 919.476.5316 Direct > +1 919.522.3348 Mobile > brianc(a)netapp.com > > > > >

1 0

diagnosing iSCSI performance on FAS2050
by John Stoffel 01 Oct '10

01 Oct '10

Hi all, I'm working to diagnose some problems with performance of a Lotus Notes server talking over iSCSI to a FAS2050 running 7.2.6.1P8 and figuring out what my options are. The system has 20 X289 420Gb SATA/SAS disks on it currently, and we are looking to add some FC disks in a new shelf. I'm not seeing any network collisions or anything like that, so I don't think it's a hardware problem, but more of a resource problem. There is some CIFS on the system in use as well, but it's mostly iSCSI traffic. Looking at the info, I don't see any terrible numbers, and the performance looks ok. Not great, but ok. Here's some output from 'statit' which I left running for about 2 minutes or so. I can go longer if need be obviously. Hostname: irvfs1 ID: 0135054398 Memory: 1792 MB NetApp Release 7.2.6.1P8: Fri Jul 17 09:03:14 PDT 2009 <8O> Start time: Fri Oct 1 11:30:24 PDT 2010 CPU Statistics 316.694026 time (seconds) 100 % 74.098481 system time 23 % 4.159199 rupt time 1 % (905915 rupts x 5 usec/rupt) 69.939282 non-rupt system time 22 % 242.595544 idle time 77 % 73.954047 time in CP 23 % 100 % 1.130218 rupt time in CP 2 % (226012 rupts x 5 usec/rupt) Miscellaneous Statistics (per second) 10966.86 hard context switches 0.00 NFS operations 125.38 CIFS operations 0.00 HTTP operations 0.00 NetCache URLs 0.00 streaming packets 2870.35 network KB received 8529.71 network KB transmitted 11154.87 disk KB read 4770.64 disk KB written 2698.64 NVRAM KB written 0.00 nolog KB written 2034.43 WAFL bufs given to clients 0.00 checksum cache hits ( 0%) 0.00 no checksum - partial buffer 0.00 FCP operations 434.87 iSCSI operations WAFL Statistics (per second) 218.61 name cache hits ( 80%) 55.35 name cache misses ( 20%) 20887.17 buf hash hits ( 82%) 4548.72 buf hash misses ( 18%) 1813.94 inode cache hits ( 100%) 6.17 inode cache misses ( 0%) 5871.88 buf cache hits ( 75%) 1963.55 buf cache misses ( 25%) 171.92 blocks read 2101.86 blocks read-ahead 384.55 chains read-ahead 88.86 dummy reads 344.25 blocks speculative read-ahead 831.96 blocks written 6.58 stripes written 0.00 blocks over-written 0.08 wafl_timer generated CP 0.00 snapshot generated CP 0.00 wafl_avail_bufs generated CP 0.00 dirty_blk_cnt generated CP 0.00 full NV-log generated CP 0.00 back-to-back CP 0.00 flush generated CP 0.00 sync generated CP 0.00 wafl_avail_vbufs generated CP 0.00 deferred back-to-back CP 0.00 low mbufs generated CP 0.00 low datavecs generated CP 1183.82 non-restart messages 121.62 IOWAIT suspends 314438 buffers RAID Statistics (per second) 312.93 xors 0.00 long dispatches [0] 0.00 long consumed [0] 0.00 long consumed hipri [0] 0.00 long low priority [0] 0.00 long high priority [0] 0.00 long monitor tics [0] 0.00 long monitor clears [0] 0.00 long dispatches [1] 0.00 long consumed [1] 0.00 long consumed hipri [1] 0.00 long low priority [1] 0.00 long high priority [1] 0.00 long monitor tics [1] 0.00 long monitor clears [1] 18 max batch 0.56 blocked mode xor 112.52 timed mode xor 0.05 fast adjustments 0.05 slow adjustments 0 avg batch start 0 avg stripe/msec 6.73 tetrises written 0.00 master tetrises 0.00 slave tetrises 179.93 stripes written 133.00 partial stripes 46.93 full stripes 832.11 blocks written 361.72 blocks read 21.89 1 blocks per stripe size 7 17.47 2 blocks per stripe size 7 16.67 3 blocks per stripe size 7 20.33 4 blocks per stripe size 7 24.39 5 blocks per stripe size 7 32.24 6 blocks per stripe size 7 46.93 7 blocks per stripe size 7 Network Interface Statistics (per second) iface side bytes packets multicasts errors collisions pkt drops e0a recv 55529.33 122.23 0.00 0.00 0.00 xmit 227177.99 228.14 0.00 0.00 0.00 e0b recv 39175.77 117.07 0.00 0.00 0.00 xmit 79516.28 126.61 0.27 0.00 0.00 e1a recv 2844533.72 2570.46 0.00 0.00 0.00 xmit 8427700.87 6604.89 0.00 0.00 0.00 e1b recv 0.40 0.01 0.00 0.00 0.00 xmit 29.44 0.27 0.27 0.00 0.00 vh recv 0.00 0.00 0.00 0.00 0.00 xmit 0.00 0.00 0.00 0.00 0.00 vif1 recv 94724.52 239.47 0.08 0.00 0.00 xmit 306742.10 354.89 0.27 0.00 0.00 iscsi recv 2863035.80 2585.27 0.02 0.00 0.00 xmit 8459234.73 6632.46 0.27 0.00 0.00 Disk Statistics (per second) ut% is the percent of time the disk was busy. xfers is the number of data-transfer commands issued per second. xfers = ureads + writes + cpreads + greads + gwrites chain is the average number of 4K blocks per command. usecs is the average disk round-trip time per 4K block. disk ut% xfers ureads--chain-usecs writes--chain-usecs cpreads-chain-usecs greads--chain-usecs gwrites-chain-usecs /aggr0/plex0/rg0: 0c.00.0 3 8.11 0.16 1.36 6206 5.20 15.58 423 2.75 6.96 377 0.00 .... . 0.00 .... . 0c.00.1 3 8.27 0.16 1.36 21309 5.38 15.13 423 2.74 7.08 402 0.00 .... . 0.00 .... . 0c.00.2 23 48.11 40.63 4.28 1733 4.37 11.31 875 3.11 8.07 616 0.00 .... . 0.00 .... . 0c.00.15 23 47.38 39.91 4.23 1708 4.27 11.16 936 3.19 8.09 544 0.00 .... . 0.00 .... . 0c.00.4 23 46.19 39.16 4.31 1739 3.89 12.76 813 3.14 7.67 631 0.00 .... . 0.00 .... . 0c.00.5 23 46.99 39.86 4.28 1761 3.98 12.16 814 3.15 8.14 529 0.00 .... . 0.00 .... . 0c.00.6 23 47.39 40.20 4.33 1711 3.95 11.92 832 3.25 8.61 542 0.00 .... . 0.00 .... . 0c.00.7 23 47.00 39.86 4.33 1738 3.92 11.96 806 3.21 8.51 567 0.00 .... . 0.00 .... . 0c.00.8 22 45.78 38.71 4.37 1688 3.87 12.57 839 3.20 7.94 563 0.00 .... . 0.00 .... . /aggr0/plex0/rg1: 0c.00.9 2 5.78 0.00 .... . 4.16 23.81 222 1.62 6.71 399 0.00 .... . 0.00 .... . 0c.00.10 2 5.78 0.00 .... . 4.18 23.68 231 1.60 6.80 384 0.00 .... . 0.00 .... . 0c.00.11 23 47.92 39.78 4.24 1740 4.77 14.60 656 3.37 8.05 550 0.00 .... . 0.00 .... . 0c.00.12 23 47.55 39.45 4.14 1860 4.79 14.57 649 3.31 8.08 514 0.00 .... . 0.00 .... . 0c.00.13 23 47.62 39.49 4.26 1725 4.94 14.52 649 3.19 7.79 532 0.00 .... . 0.00 .... . 0c.00.19 24 48.22 39.78 4.18 1849 5.07 13.97 689 3.37 7.91 512 0.00 .... . 0.00 .... . 0c.00.17 23 48.19 40.11 4.16 1811 4.82 14.66 641 3.25 8.15 553 0.00 .... . 0.00 .... . 0c.00.14 23 47.77 39.81 4.16 1834 4.74 14.92 609 3.21 8.00 530 0.00 .... . 0.00 .... . 0c.00.16 24 48.31 40.05 4.13 1829 4.98 14.28 688 3.27 7.88 584 0.00 .... . 0.00 .... . Aggregate statistics: Minimum 2 5.78 0.00 3.87 1.60 0.00 0.00 Mean 18 38.46 30.95 4.52 2.99 0.00 0.00 Maximum 24 48.31 40.63 5.38 3.37 0.00 0.00 Spares and other disks: 0c.00.3 0 0.00 0.00 .... . 0.00 .... . 0.00 .... . 0.00 .... . 0.00 .... . Spares and other disks: 0c.00.18 0 0.00 0.00 .... . 0.00 .... . 0.00 .... . 0.00 .... . 0.00 .... . FCP Statistics (per second) 0.00 FCP Bytes recv 0.00 FCP Bytes sent 0.00 FCP ops iSCSI Statistics (per second) 2672116.49 iSCSI Bytes recv 8064194.15 iSCSI Bytes xmit 434.87 iSCSI ops Interrupt Statistics (per second) 999.38 Clock (IRQ 0) 0.05 Uart (IRQ 4) 1766.58 PCI direct (IRQ 16) 94.52 PCI direct (IRQ 17) 0.00 RTC 2860.54 total NVRAM Statistics (per second) 0.00 total dma transfer KB 0.00 wafl write req data KB 0.00 dma transactions 0.00 dma destriptors 28.36 transactions not queued 67.01 transactions queued 12.38 total waittime (MS) 29.34 completion proc wakeups 18.39 completion proc waitdone 186.57 waitdone preempts 0.00 waitdone delays 2698.66 total nvlog KB 0.00 nvlog shadow header array full 0.00 channel1 dma transfer KB 0.00 channel1 dma transactions 0.00 channel1 dma descriptors E7520 Data Mover Statistics (per second) 3810.35 total dma transfer KB 1.18 total bcopy transfer KB 4.22 total waittime (MS) I've also got some 'sysstat -i 5' numbers: > sysstat -i 5 CPU NFS CIFS iSCSI Net kB/s Disk kB/s iSCSI kB/s Cache in out read write in out age 15% 0 195 247 1232 4944 7678 4267 1104 4629 1 16% 0 238 352 1959 6185 7414 3201 1770 4791 1 11% 0 135 264 1908 4533 4946 0 1770 4257 1 18% 0 171 246 1155 4372 6915 7230 1046 4084 1 14% 0 230 298 1868 4478 4892 0 1717 4184 2 17% 0 164 284 1113 4955 7014 5319 1001 4683 2 9% 0 18 238 1395 3836 4217 0 1303 3654 2 15% 0 12 245 1759 4110 7539 3378 1653 3912 3 13% 0 21 244 900 5029 6302 3205 824 4814 2 11% 0 22 282 1569 5152 5866 0 1461 4910 2 19% 0 21 389 2280 5927 9750 6982 2132 5644 2 14% 0 49 401 1746 6072 6506 0 1614 5775 3 27% 0 83 491 5246 8447 12100 7693 4961 7985 2 11% 0 126 263 1772 4800 5258 0 1642 4553 2 16% 0 20 487 802 7104 10182 718 694 6800 6 25% 0 15 518 1763 6440 10771 9322 1621 6144 2 15% 0 81 389 1745 6271 6844 0 1607 5979 3 20% 0 140 287 906 6280 10088 7681 782 5965 4 8% 0 70 201 1163 4381 4638 0 937 4148 2 Any thoughts? Thnaks, John John Stoffel - Senior Staff Systems Administrator - System LSI Group Toshiba America Electronic Components, Inc. - http://www.toshiba.com/taec john.stoffel(a)taec.toshiba.com - 508-486-1087

2 1

What do you monitor
by Nicholas Bernstein 30 Sep '10

30 Sep '10

I had an interesting question posed by a student yesterday - "aside from interfaces being up, and volumes being online, what do you typically monitor?" Some of my initial thoughts were: - inodes being available - ifstat on multimode vifs to make sure interfaces are being used - io counts on luns/qtrees/vols/ - client side nfsstat or the equivalent for that protocol from the client side Anyway, I was trying to think of some of the "non-standard" things to monitor, and thought I'd put it out to the list and see what other people are typically doing. Cheers, Nick

3 2

Switching from single network interface to vif live?
by Phelps, Matthew 30 Sep '10

30 Sep '10

Has anyone switched from a single network interface to a vif on a running system? I'd like to avoid a reboot if possible. -- Matt Phelps System Administrator, Computation Facility Harvard - Smithsonian Center for Astrophysics mphelps(a)cfa.harvard.edu, http://www.cfa.harvard.edu

2 1

Second interface up, routed on
by Kevin Sheen 23 Sep '10

23 Sep '10

Hi, We recently brought up a second interface on our 3070 cluster. I was double checking /vol0/etc/rc on both filers and the interfaces are in there for when we reboot. I also saw that routed was set to on. Should I disable routed? I took the Netapp Fundamentals class over 3 years ago and can't remember if they said it should be disabled. I don't really want our filers to advertise or become a router for the network. If I disable routed, will it require a reboot to take effect? thanks, Kevin

4 3

how do you put "ip.tcp" options into effect?
by Fred Grieco 16 Sep '10

16 Sep '10

Do I need to do anything for "ip.tcp.newreno.enable" and "ip.tcp.sack.enable" options changes to take effect? Is it automatic, or do i need to reboot? Thanks, Fred

1 0

Re: CIFS access question
by Page, Jeremy 11 Sep '10

11 Sep '10

If you are a Windows guy I think you can use the computer management MMC. It also is much nicer than the insanity you get with a "cifs sessions" on a busy filer. You'd think a bunch of Sun and BSD guys would understand why stuff like regex. Please use ^D to delete this post... Romeo Theriault <romeotheriault(a)gmail.com> wrote: > > Any hints appreciated. Hi John, If you want to configure local user accounts on the filer for your cifs users, when you run 'cifs setup' make sure you choose option 3, "Windows Workgroup Authentication". Then you can create a user account on the filer for that cifs user. useradmin group add cifsonly -c "This group is for non-admin CIFS access" -r none useradmin user add appuser -c "Non-administrative CIFS-only user" -g cifsonly Now, because of the way that the filers map windows id's to unix id's you still need to create an entry in the filers /etc/passwd file for the user you just created. Give it a unique uid and gid that are higher than the uid's and gid's that are currently in the /etc/passwd file and just give it a '/' as the home directory. Example: root:_K9..qPPDSr/2SCvwCSTSc:0:1::/: pcuser::65534:65534::/: nobody::65535:65535::/: appuser::65536:65536::/: In my case, due to permissions issues, I needed to map my cifs user to root to get around a permission's issue. You probably don't need it but here is the syntax I used to do so. *\appuser => root Now you should be able to create a share, assign the appuser privs to it and they should be able to get in. If not you can turn on cifs login tracing like so: options cifs.trace_login on Good luck! -- Romeo Theriault System Administrator Information Technology Services Please be advised that this email may contain confidential information. If you are not the intended recipient, please notify us by email by replying to the sender and delete this message. The sender disclaims that the content of this email constitutes an offer to enter into, or the acceptance of, any agreement; provided that the foregoing does not invalidate the binding effect of any digital or other electronic reproduction of a manual signature that is included in any attachment.

1 0

CIFS access question
by John Stoffel 10 Sep '10

10 Sep '10

Hi all, I'm a Unix/NFS guy, but I've been asked to allow CIFS access from a Windows box to a volume currently mounted by NFS to a Solaris box. Not a problem, I'll just use our existing CIFS setup and setup a share for that volume. The problems are in terms of authentication. I'd like to restrict the CIFS share for /vol/appvol to JUST this single windows host (winbox) and user (appuser). But the user "appuser" is a purely local account on both the Windows and Unix systems. They're the same name luckily, but neither is on the NIS or AD domains. So I've setup the usermap.cfg on the Netapp (3140, OnTap 7.3.1.1) to look like this: winbox\#appuser <= appuser But I've also tried: winbox\appuser == appuser *\appuser == appuser *\appuser => appuser And I've setup the share with: rsh filer cifs share access appvol appuser Full Control but without any luck. It still asks for a password, but then refuses to let me in. Note that the filer doesn't really have a concept of this username, since it's purely local to the Unix and Window hosts. Do I need to add this username into NIS and/or AD to make things work? So I decided to try first by adding the 'appuser' account into my NIS domain, with a matching password to the host local Windows account. No luck so far. Looking at things using 'wcc -u appuser' on the filer, I get "Mapped user not found" even though I can now see that userin NIS on other unix clients. The funky thing is that I can see with 'cifs sessions' that there is a CIFS connection from the Windows box to the filer, but it says: x.y.z.2(WINBOX) is connected, but has no users. so something is working, but not the authentication. I see in the messages file on the netapp the following: Fri Sep 10 10:34:01 EDT [auth.trace.authenticateUser.loginTraceIP:info]: AUTH: Login attempt by user appuser of domain WINBOX from client machine x.y.z.2 (WINBOX). Fri Sep 10 10:34:01 EDT [auth.dc.trace.DCConnection.statusMsg:info]: AUTH: Trace DC- attempting authentication with domain controller \\MARL-DC. Fri Sep 10 10:34:01 EDT [auth.trace.authenticateUser.loginRejected:info]: AUTH: Login attempt by user rejected by the domain controller with error 0xc0000064: D C indicates user is not from a trusted domain. Fri Sep 10 10:34:01 EDT [auth.trace.authenticateUser.loginTraceMsg:info]: AUTH: login from x.y.z.2 rejected because guest account not set. Fri Sep 10 10:34:01 EDT [auth.trace.authenticateUser.loginTraceMsg:info]: AUTH: Delaying the response by 5 seconds due to continuous failed login attempts by us er appuser of domain WINBOX from client machine x.y.z.2. So maybe I need to add in a guest user? But I don't want a general guest user at all, I really want a very limited share setup for just two hosts and one username. Any hints appreciated. Thanks, John John Stoffel - Senior Staff Systems Administrator - System LSI Group Toshiba America Electronic Components, Inc. - http://www.toshiba.com/taec john.stoffel(a)taec.toshiba.com - 508-486-1087

2 1

RE: CIFS access question
by John Stoffel 10 Sep '10

10 Sep '10

>>>>> "steve" == steve klise <sklise(a)hotmail.com> writes: steve> Can the user \\filer\appshare and see the share? No, it keeps asking for a username/password. steve> If not, that would be a problem, and the share is not setup, or steve> visible. I can browse \\filer from another client (Vista) and see the share (one of a couple on this filer) but I cannot access it, which is ok from that client. Hmm.. when I try to map the \\filer\appvol to my laptop, using the appuser account and password, it doesn't let me on. Which is ok, but ideally should be allowed. Esp since I now have: *\appuser => appuser in the usermap.cfg file, and the password on the Unix side is known. steve> You need to create a local user on the filer with the same steve> password, or add the domain or local user to the NTFS security, steve> and I would add the computerobject account. I don't have any privs to add users to the domain, nor do we want to add this user to the domain, it's a purely local account to the Windows box. Since I'm a unix admin, it's easy enough for me to add the Unix account to NIS for testing, though ideally it would be host local as well. Adding it to the Netapp is possible as well. steve> Usually I do this via the MMC, but that is maybe not an option steve> for you. MMC? I could use the fileradmin web page stuff... Thanks for the quick reply, John John Stoffel - Senior Staff Systems Administrator - System LSI Group Toshiba America Electronic Components, Inc. - http://www.toshiba.com/taec john.stoffel(a)taec.toshiba.com - 508-486-1087 >> Date: Fri, 10 Sep 2010 10:39:00 -0400 >> From: john.stoffel(a)taec.toshiba.com >> To: toasters(a)mathworks.com >> Subject: CIFS access question >> >> >> Hi all, >> >> I'm a Unix/NFS guy, but I've been asked to allow CIFS access from a >> Windows box to a volume currently mounted by NFS to a Solaris box. >> Not a problem, I'll just use our existing CIFS setup and setup a share >> for that volume. >> >> The problems are in terms of authentication. I'd like to restrict the >> CIFS share for /vol/appvol to JUST this single windows host (winbox) >> and user (appuser). >> >> But the user "appuser" is a purely local account on both the Windows >> and Unix systems. They're the same name luckily, but neither is on >> the NIS or AD domains. >> >> So I've setup the usermap.cfg on the Netapp (3140, OnTap 7.3.1.1) to >> look like this: >> >> winbox\#appuser <= appuser >> >> But I've also tried: >> >> winbox\appuser == appuser >> *\appuser == appuser >> *\appuser => appuser >> >> >> And I've setup the share with: >> >> rsh filer cifs share access appvol appuser Full Control >> >> but without any luck. It still asks for a password, but then refuses >> to let me in. Note that the filer doesn't really have a concept of >> this username, since it's purely local to the Unix and Window hosts. >> Do I need to add this username into NIS and/or AD to make things work? >> >> So I decided to try first by adding the 'appuser' account into my NIS >> domain, with a matching password to the host local Windows account. >> No luck so far. Looking at things using 'wcc -u appuser' on the >> filer, I get "Mapped user not found" even though I can now see that >> userin NIS on other unix clients. >> >> The funky thing is that I can see with 'cifs sessions' that there is a >> CIFS connection from the Windows box to the filer, but it says: >> >> x.y.z.2(WINBOX) is connected, but has no users. >> >> so something is working, but not the authentication. I see in the >> messages file on the netapp the following: >> >> Fri Sep 10 10:34:01 EDT [auth.trace.authenticateUser.loginTraceIP:info]: AUTH: Login attempt by user appuser of domain WINBOX from client machine x.y.z.2 (WINBOX). >> Fri Sep 10 10:34:01 EDT [auth.dc.trace.DCConnection.statusMsg:info]: AUTH: Trace >> DC- attempting authentication with domain controller \\MARL-DC. >> Fri Sep 10 10:34:01 EDT [auth.trace.authenticateUser.loginRejected:info]: AUTH: >> Login attempt by user rejected by the domain controller with error 0xc0000064: D >> C indicates user is not from a trusted domain. >> Fri Sep 10 10:34:01 EDT [auth.trace.authenticateUser.loginTraceMsg:info]: AUTH: >> login from x.y.z.2 rejected because guest account not set. >> Fri Sep 10 10:34:01 EDT [auth.trace.authenticateUser.loginTraceMsg:info]: AUTH: >> Delaying the response by 5 seconds due to continuous failed login attempts by us >> er appuser of domain WINBOX from client machine x.y.z.2. >> >> >> So maybe I need to add in a guest user? But I don't want a general >> guest user at all, I really want a very limited share setup for just >> two hosts and one username. >> >> Any hints appreciated. >> >> Thanks, >> John >> John Stoffel - Senior Staff Systems Administrator - System LSI Group >> Toshiba America Electronic Components, Inc. - http://www.toshiba.com/taec >> john.stoffel(a)taec.toshiba.com - 508-486-1087 steve> <html> steve> <head> steve> <style> </style> steve> </head> steve> <body class='hmmessage'> steve> Can the user <A href="file://\\filer\appshare">\\filer\appshare</A> and see the share? steve>   steve> If not, that would be a problem, and the share is not setup, or visible. steve>   steve> You need to create a local user on the filer with the same password, or add the domain or local user to the NTFS security, and I would add the computerobject account.  steve>   steve> Usually I do this via the MMC, but that is maybe not an option for you.    steve> > Date: Fri, 10 Sep 2010 10:39:00 -0400 > From: john.stoffel(a)taec.toshiba.com > To: toasters(a)mathworks.com > Subject: CIFS access question > > > Hi all, > > I'm a Unix/NFS guy, but I've been asked to allow CIFS access from a > Windows box to a volume currently mounted by NFS to a Solaris box. > Not a problem, I'll just use our existing CIFS setup and setup a share > for that volume. > > The problems are in terms of authentication. I'd like to restrict the > CIFS share for /vol/appvol to JUST this single windows host (winbox) > and user (appuser). > > But the user "appuser" is a purely local account on both the Windows > and Unix systems. They're the same name luckily, but neither is on > the NIS or AD domains. > > So I've setup the usermap.cfg on the Netapp (3140, OnTap 7.3.1.1) to > look like this: > > winbox\#appuser <= appuser > > But I've also tried: > > winbox\appuser == appuser > *\appuser == appuser > *\appuser => appuser > > > And I've setup the share with: > > rsh filer cifs share access appvol appuser Full Control > > but without any luck. It still asks for a password, but then refuses > to let me in. Note that the filer doesn't really have a concept of > this username, since it's purely local to the Unix and Window hosts. > Do I need to add this username into NIS and/or AD to make things work? > > So I decided to try first by adding the 'appuser' account into my NIS > domain, with a matching password to the host local Windows account. > No luck so far. Looking at things using 'wcc -u appuser' on the > filer, I get "Mapped user not found" even though I can now see that > userin NIS on other unix clients. > > The funky thing is that I can see with 'cifs sessions' that there is a<BR! >> CI F S connection from the Windows box to the filer, but it says: > > x.y.z.2(WINBOX) is connected, but has no users. > > so something is working, but not the authentication. I see in the > messages file on the netapp the following: > > Fri Sep 10 10:34:01 EDT [auth.trace.authenticateUser.loginTraceIP:info]: AUTH: Login attempt by user appuser of domain WINBOX from client machine x.y.z.2 (WINBOX). > Fri Sep 10 10:34:01 EDT [auth.dc.trace.DCConnection.statusMsg:info]: AUTH: Trace > DC- attempting authentication with domain controller \\MARL-DC. > Fri Sep 10 10:34:01 EDT [auth.trace.authenticateUser.loginRejected:info]: AUTH: > Login attempt by user rejected by the domain controller with error 0xc0000064: D > C indicates user is not from a trusted domain. > Fri Sep 10 10:34:01 EDT [auth.trace.authenticateUser.loginTraceMsg:info]: AUTH: > login from x.y.z.2 rejected because guest account not set. > Fri Sep 10 10:34:01 EDT [auth.trace.authenticateUser.loginTraceMsg:info]: AUTH: > Delaying the response by 5 seconds due to continuous failed login attempts by us > er appuser of domain WINBOX from client machine x.y.z.2. > > > So maybe I need to add in a guest user? But I don't want a general > guest user at all, I really want a very limited share setup for just > two hosts and one username. > > Any hints appreciated. > > Thanks, > John > John Stoffel - Senior Staff Systems Administrator - System LSI Group > Toshiba America Electronic Components, Inc. - http://www.toshiba.com/taec > john.stoffel(a)taec.toshiba.com - 508-486-1087 </body> steve> </html>

1 0

Netapp with ZenDesktop or View 4.5
by steve klise 08 Sep '10

08 Sep '10

Just wondering what folks are doing for making a decesion for View 4.5 (yeah, I know it just came out) or zendesktop. I was at VMWorld and I am drinking the View cool-aid. I know in the past vdi could deploy thousands of vdi machines with flexclone, thin provisioning, etc on Netapps. We have a 6040 with a PAM card and we use ESX 4.x over FC on it (DOT7.3.2x). No flexclone license on it,but I will on a 6080 (DOT 7.3.2.x) with a PAMII card. Couple questions. Are the dedupe numbers the same for both solutions (VM vs Citrix) ? Is the solution a moot point, or would one be better from a performance, and or dedupe savings perspective? Most likely we will have the storage between 2 HA heads, so we try to design redundancy as much as possible. I was hoping to serve up 2000 users on a SAS fully populated shelf on a 6040 and another shelf on a 6080 for 2k users each. These are just rough numbers and I will read the best practices for each solution. Just looking for any feedback. -- View this message in context: http://old.nabble.com/Netapp-with-ZenDesktop-or-View-4.5-tp29643248p2964324… Sent from the Network Appliance - Toasters mailing list archive at Nabble.com.

1 0

Jump to page:

2025

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