toasters

toasters@lists.teaparty.net

13531 discussions

Re: Multiple IP Addresses per interface?
by Keith Brown 17 Feb '98

17 Feb '98

From: Brian Tao <taob(a)nbc.netcom.ca> > > According to the man page: > > The interface name vh is used to specify IP virtual host > addresses associated with the filer. Only alias addresses (using > the alias option) may be assigned to the vh interface. > > So I can assign multiple IP's to vh... how do I bind those IP's to >a physical interface? Or is this only used for the virtual hosting >feature in the HTTP server? Unfortunately the doc on this subject doesn't really spell all this out in a manner that could be described as crystal clear, but... My understanding is that the vh interface is just associated with the HTTP daemon's virtual hosting feature. The addresses on vh are the ones that the web server runs through /etc/httpd.hostprefixes to locate a target specific home directory for inbound HTTP requests. However, there is of course the "little" issue of packets destined for a vh address being able to find their way into a filer in the first place, which will inevitably require some cooperation from the real, physical interfaces. This could conceivably be made to happen via routing trickery, perhaps using host routes to ensure that packets destined for vh addresses get fired at a filers physical interfaces on their last hop, or maybe even full network routes, where the vh addresses are on a subnet that is entirely internal to the filer and the physical interfaces form routes to that subnet. Failing this, then you're right, there needs to be some correlation between vh addresses and the addresses being supported by the physical interfaces. Alas I haven't been able to find a description of whether this is a manual or automatic process. I have to assume manual for now. My own experience starts and ends with loading up just the physical interfaces with addresses, and leaving the vh interface alone. Keith

1 0

NetApp 450 end of life
by Mark Richards 17 Feb '98

17 Feb '98

Hello We have a Netapp 450 (the 486 EISA-based machine). NetApp will soon discontinue support for the machine and I'm curious as to what others think of this. We bought our 450 in Nov. 1995 before the PCI-based filers appeared. We are a fairly small shop, (nothing like some of the huge servers discussed on this list :-), but we are delighted with our filer which has never crashed or given us any problems. I understand NetApp reasoning for desupporting the 450 - it costs money to test each new version of the OS on the EISA boxes for what I assume is a small number of remaining users. What do other successful vendors do with older hardware? Does the newest Solaris run on older Sparcs? What about CISCO? I have some seven year old Telebit netblazer routers and they still run the newest OS. NetApp compares its filers to routers, why can't they do what Telebit has been able to do? Our 450 is 27 months old....its performance is terrific for our purposes and I'd have a hard time forming a technical and business case to replace the system with a new model. I guess I just don't think 27 months is VERY OLD! What do you think? I'm not bashing NetApp, they're a great company, I'm just disappointed. Regards Mark Richards Hicks Morley Hamilton Stewart Storie

7 6

None
by Ron Thibault 17 Feb '98

17 Feb '98

subscribe ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Ron Thibault 2820 San Tomas Expressway Network Appliance Santa Clara, CA. 95051 Customer Support Rep. Voice: (408) 367-3742 www.netapp.com Fax: (408) 367-5151 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ "The best thing in life is definitely free." - by Big Tent Revival

1 0

None
by Manoj Gandhi 17 Feb '98

17 Feb '98

subscribe ) ========================================================= Manoj Gandhi Customer Support Representative Network Appliance (408)367-3198 http://www.netapp.com/ _________________________________________________________

1 0

Exabyte Mammoth (exb-8900) and 220 changer support?
by Dave Cole 16 Feb '98

16 Feb '98

I am in the process of evaluating tape units and came upon the startling discovery that (as of at least) ontap 4.3R3, there is still no support for the Exabyte Mammoth tape drive. Does anyone know any details about this issue? I would really like to be able to recommend the Exabyte Mammoth 20/40GB tape drives. ------------------------------------------------------------- Dave Cole (DC1110) | dacole(a)netcom.ca Systems Administrator | dacole(a)vex.net | office/~dacole/ Netcom Canada | www.vex.net/~dacole/ 905 King Street West, Toronto, M6K 3G9 | phone - 416.341.5801 Toronto, Ontario, Canada, Earth, Sol | fax - 416.341.5725

1 1

Re: Multiple IP Addresses per interface?
by Keith Brown 16 Feb '98

16 Feb '98

>The purpose of this email is to ask if I can configure multiple IP >addresses per physical interface on an F330. Sure. It's very easy using the "alias" option to ifconfig(1). For example, the sequence: ifconfig e0 192.1.1.1 ifconfig e0 alias 192.1.1.2 will put a couple of addresses on the e0 interface. >If it is possible I did >not see reference to it in the manuals. It's described on the na_ifconfig(1) man page in the command reference guide for Data ONTAP 4.3.X. If you have a NOW account, you can obtain this electronically from: http://now.netapp.com/knowledge/docs/crq41.pdf >the filer is a F330 running 4.0.2c (?) and I have NFS over TCP turned off. Hmmm. I don't know offhand when the multiple IP addresses on a single interface support was formally introduced. There's a chance it was post 4.0.2c, although I'm not certain. If your na_ifconfig(1) man page doesn't mention the "alias" option, you may need to get a hold of a newer software release. Again, you can download that from the NOW web site if you have an account there. Keith

2 1

GNU find patch to fix .snapshot annoyance
by Daniel Quinlan 16 Feb '98

16 Feb '98

It's really annoying when find(1) descends into a .snapshot directory. Here's a patch against GNU findutils 4.1 tested under Linux and Solaris to fix the problem. It still allows descent into .snapshot, but only if .snapshot is specifically listed on the command line. GNU findutils is available at <URL:ftp://prep.ai.mit.edu/pub/gnu>. Dan --- findutils-4.1/find/find.c~ Wed Oct 12 14:21:11 1994 +++ findutils-4.1/find/find.c Sun Feb 15 17:21:29 1998 @@ -38,6 +38,10 @@ #define lstat stat #endif +#ifndef S_IRWXUGO +#define S_IRWXUGO (S_IRWXU|S_IRWXG|S_IRWXO) +#endif + int lstat (); int stat (); @@ -390,8 +394,11 @@ apply_predicate (pathname, &stat_buf, eval_tree); if (stop_at_current_level == false) - /* Scan directory on disk. */ - process_dir (pathname, name, strlen (pathname), &stat_buf, parent); + /* Skip over Network Appliance ".snapshot" directories. */ + if (!(!strcmp(name, ".snapshot") && (stat_buf.st_uid == 0) && + (stat_buf.st_gid == 0) && ((stat_buf.st_mode & 0777) == S_IRWXUGO))) + /* Scan directory on disk. */ + process_dir (pathname, name, strlen (pathname), &stat_buf, parent); if (do_dir_first == false && curdepth >= mindepth) apply_predicate (pathname, &stat_buf, eval_tree);

1 0

Multiple IP Addresses per interface?
by David Drum 15 Feb '98

15 Feb '98

Hello everyone, The purpose of this email is to ask if I can configure multiple IP addresses per physical interface on an F330. If it is possible I did not see reference to it in the manuals. The reason I am asking about this may be of interest to some, though, so an explanation follows. Upon installing a Cisco Catalyst network switch, communication between the filer and our staff machine went bad. I am the only one using the filer from the staff machine so I have only researched the problem as time allows. The staff machine is a Sun Ultra 2 running Solaris 2.5.1, the filer is a F330 running 4.0.2c (?) and I have NFS over TCP turned off. The first thing I did was have our network guys manually configure all the ports on the Catalyst, since this list had discussed the general problem with ethernet autonegotiation. It did not help. The next thing I did was replace the ethernet cable from the staff machine to the Catalyst. If the staff machine was the only device having a problem, it stood to reason. No effect. Next, I began a long series of large packet pings between various machines connected to the Catalyst. As a result of that, I have discovered that a Catalyst does not like multiple subnets and drops from 12 to 22 percent of cross-subnet packets. My network guru claims that the Catalyst switches at layer 2 and so the IP address should not matter, but that certainly is at odds with what I am seeing. I'll let the list know how this turns out OR if you have a Catalyst serving multiple subnets without problem, please let me know how it is configured. Just for the record, the filer and its two main clients are on three successive ports on the Catalyst. Regards, David K. Drum david(a)more.net -- It's hard to be bored when you're as stupid as a line. [1] Reality has a tendency to be so uncomfortably real. [2] You can only measure the size of your head from the inside. [3] When you proceed deliberately, mistakes don't cascade, they instruct. [4] [1] Vernor Vinge [2] Neil Peart [3] Larry Wall [4] Stewart Brand

1 0

[Fwd: Exabyte Mammoth (exb-8900) and 220 changer support?]
by Steve Gremban 13 Feb '98

13 Feb '98

Dave, In the February '98 issue of SysAdmin magazine they do an evaluation of the Sony AIT, Exabyte Mammoth, and the Quantum DLT. They compare head life, tape capacity, cost per GB compressed/uncompressed, media load and file access times, transfer rates compressed/uncompressed, etc., and end up recommending the Sony. I am using Quantum DLT technology and find it performs well, is reliable, and is supported on the Netapp. I'm not sure whether the Sony or Exabyte are supported yet. -Steve Gremban gremban(a)ti.com

2 1

subscribe
by Mark Conley 12 Feb '98

12 Feb '98

1 0

Re: Moving disks between filers
by Robert Hamilton 11 Feb '98

11 Feb '98

Amy, You can install the F220 disks in any available SCSI slot with a SCSI ID between 0 and 6. In the factory default settings on the F540 "shelf pair" these are the remaining 6 slots in the shelf that has the differential converter installed in the far right position. The F220 drives are narrow (8-bit), so they can't be recognized in SCSI IDs >7. A lot of our customers are using recycled F220/F330 disk drives in their newer systems, including the F540, F520 and F630. Regards, Robert Hamilton NetApp >At 10:05 AM 2/11/98 -0500, Amy Chused wrote: >>I have a 540 and a 220. The 220 has been taken out of production and is >>going to be reinitialized before it is used again. The 220 disk shelves >>are full (lots o' disk). >> >>The 540 is in production, and isn't short of disk space right now, but we >>all know how swiftly disk space vanishes. >> >>I'd like to know if it's possible to use disks from the 220 in the 540 disk >>shelves. Does anyone know? >> >>Thanks, >> >> Amy >> >> > >Respectfully, >Greg Collins > > > NETWORK APPLIANCE >========================================================================= > FAST! SIMPLE! RELIABLE! MULTIPROTOCOL!! >========================================================================= >Greg Collins support: 1-888-4NETAPP >District Manager/MidAtlantic support email: support(a)netapp.com >voice: 301-230-5840 main: 1-800-952-5005 >fax: 301-230-5852 URL: http://www.netapp.com/ >email: gcollins(a)netapp.com >pager: 1-800-891-2746 >pager-mail: gcollins-pager(a)netapp.com > >=========================================================================== > > >

1 0

Moving disks between filers
by Amy Chused 11 Feb '98

11 Feb '98

I have a 540 and a 220. The 220 has been taken out of production and is going to be reinitialized before it is used again. The 220 disk shelves are full (lots o' disk). The 540 is in production, and isn't short of disk space right now, but we all know how swiftly disk space vanishes. I'd like to know if it's possible to use disks from the 220 in the 540 disk shelves. Does anyone know? Thanks, Amy

1 0

Logging in to CIFS from W95 using multiple loginids
by Steve Losen 10 Feb '98

10 Feb '98

One of our staff is working on an application that makes it easy to map a netapp share to a drive letter. This application works fine on NT, but W95 is causing trouble. The application is a little visual basic program that opens up a box that allows the user to enter a loginid, password, and shows a list of drive letters to map to. The user sets these things and the application supplies the name of the share by building it from our NetApp's name and the user's loginid, eg, \\home1\abc9d. On NT this works great. You can specify different loginids and passwords for different shares and map them to various drive letters. On W95, however, this doesn't work so well. Apparently the calls that the visual basic app makes are replacing the loginid that the user specifies with the loginid that the user logged in to W95 with. In other words, this app cannot override the current W95 loginid. If you specify the same loginid that you logged in to W95, then it works fine. We are curious if anyone else has any experience with this problem and has any suggestions for fixing it. For the curious, we want an application like this because our NetApp box is going to eventually have around 20,000 CIFS accounts and we are going to use the unix passwd authentication rather than set up a NT domain controller and have to pay license fees per user, etc. We don't plan to have any WINS servers know about the NetApp. We have too many different networks at this University to rely on every net being able to browse for the NetApp in Network Neighborhood, etc. Instead, we are going to hand out this little app that knows the name of the netapp, making browsing, and/or remembering the name of the server unnecessary. Plus, this app will disconnect after a certain amount of idle time. Because various departments maintain their own NT file servers (with their own loginids), we can't assume that a user will always be able to login to W95 using the same loginid that they have on the netapp, hence the need to override the W95 loginid. Also, in some of our lab environments, each PC automatically logs in to W95 on boot using a "dummy" machine loginid. Has anyone done anything like this? If we can't override the W95 loginid, could our application somehow change it, connect to the server, and change it back? Steve Losen scl(a)virginia.edu phone: 804-924-0640 University of Virginia ITC Unix Support

1 0

Re: More capacity on lower-end filers
by hitz＠netapp.com 10 Feb '98

10 Feb '98

> On Thu, 12 Jun 1997, Dave Hitz wrote: > > > > Although marketing certainly has input, the space restrictions also > > come from engineering droids. During normal operation the smaller > > machines could certainly handle more disk, but the time required for > > something like RAID reconstruction could get dangerously long. > > Is this the reason why the F210 can still only use around 50GB of > disk, even though it can physically accomodate more using 9GB drives? In order for a RAID reconstruction to complete, you have to read all of the data on all of the other disks. So RAID reconstruction time is proportional the the amount of data, not proportional to the number of disks. Dave Hitz hitz(a)netapp.com Network Appliance (408) 367-3106

8 10

Unexpected behaviour from F330 running 4.0.1c and Solaris 2.5.1
by Christoph Doerbeck 09 Feb '98

09 Feb '98

Greetings, we've isolated our little problem but... the behaviour or the solaris system seemed a bit unexpected to me. There exhisted a >2GB file on our NAS and when the Solaris box encountered it, all communication (NFS) to the toaster became locked. We could ping, telnet etc... but no NFS. Once we tracked down the culprit, removing the file promptly returned the system to normal. If I understand things correctly, Solaris 2.6 now allows larger files >4GB. Anyway, I thought I'd post as a informational blurb, and maybe someone could outline the latest file (system) limits for the latest versions of the popular operating systems (HP,SGI,AIX, ..er...cough....gag...NT, and others) Maybe this would be a good QA for the NetApp FAQ some of you discussed a few months ago. - Christoph

1 0

archive is now searchable... ish.
by Tom Yates 09 Feb '98

09 Feb '98

The archive can now be searched through a simple search engine. Note that this search engine, like the web server itself, runs on my desktop machine. If I find my work going very slowly because you're all searching the archives like crazed weasels, I'm likely to kick back. Please be nice. The search engine can be found at http://teaparty.mathworks.com:1999/toasters/Search.html . In case anyone wasn't aware, the list is archived at http://teaparty.mathworks.com:1999/toasters/ . There is a reference to the search engine in the "about this archive" page, which is referenced at the very very top of the main archive page, in case you forget where the engine is. If people find this search engine not working, or think it could be improved in ways that won't take me a week, please let me know. Tom Yates - Unix Chap - The Mathworks, Inc. - +1 (508) 647 7561 MAG#65061 DoD#0135 AMA#461546 1024/CFDFDE39 0C E7 46 60 BB 96 87 05 04 BD FB F8 BB 20 C1 8C

1 0

Apache/Netapp problem + solution (fwd)
by David R. Van Sandt 05 Feb '98

05 Feb '98

What version of ONTAP are you using on the NetApp? ---------------------------------------------------------------- David R. Van Sandt, dv(a)corp.earthlink.net, 626/296-5137 Earthlink Network, Lead System Administrator "It's Your Internet" ---------------------------------------------------------------- ---------- Forwarded message ---------- Date: Wed, 4 Feb 1998 13:39:14 -0500 (EST) From: rasmus(a)bellglobal.com To: toasters(a)mathworks.com Subject: Apache/Netapp problem + solution Apache 1.2.5 on a Solaris 2.5.1 Sparc box connected to a Netapp 630 filer. Problem symptom: Web server established a connection, but would not respond to requests. Looking at the parent httpd process we see that it is in a normal "wait for something to do" loop: 0.000000 sigsuspend([] <unfinished ...> 0.415977 --- SIGALRM (Alarm Clock) --- 0.000214 <... sigsuspend resumed> ) = -1 EINTR (Interrupted system call) <0.415636> 0.000272 setcontext({uc_sigmask=[ALRM], ...}) = ? <0.000113> 0.000525 alarm(0) = 0 <0.000083> 0.000389 sigprocmask(SIG_UNBLOCK, [ALRM], NULL) = 0 <0.000087> 0.000475 sigaction(SIGALRM, {SIG_DFL}, NULL) = 0 <0.000086> 0.000464 waitid(P_ALL, 0, {si_signo=0, si_code=SI_USER, si_pid=0, si_uid=0, ...}, WNOHANG|WEXITED|WTRAPPED) = 0 <0.000091> 0.000499 alarm(0) = 0 <0.000084> 0.000349 sigaction(SIGALRM, {0xef5b8cfc, [], 0}, {SIG_DFL}) = 0 <0.000094> 0.000570 sigprocmask(SIG_BLOCK, [ALRM], []) = 0 <0.000090> 0.000577 alarm(1) = 0 <0.000086> 0.000351 sigsuspend([] <unfinished ...> 0.995342 --- SIGALRM (Alarm Clock) --- 0.000184 <... sigsuspend resumed> ) = -1 EINTR (Interrupted system call) <0.995330> 0.000248 setcontext({uc_sigmask=[ALRM], ...}) = ? <0.000107> 0.000517 alarm(0) = 0 <0.000083> Looking at the spawned child httpd's we see: 0.000000 fcntl(21, F_SETLKW, {type=F_WRLCK, whence=SEEK_SET, start=0, len=0} And they got there by: ef5b6fd8 fcntl (15, 7, 5cfcc) ef5b6fd8 _libc_fcntl (15, 7, 5cfcc, 0, 0, 0) + 8 ef78aba8 s_fcntl (15, 7, 5cfcc, ef611938, 34, effff944) + 164 00019138 accept_mutex_on (62d08, 1, 0, effffa78, effffa88, 2) + 18 0001a874 child_main (6c170, 5d000, 5d000, 5cc00, 5d000, 5b000) + 1f8 0001ad1c make_child (61880, 2, 61400, 11, 0, 5cfc9) + d4 0001b3a0 standalone_main (effffb78, 5cf48, 5f400, 5cc00, 75, 5f7b0) + 2a4 0001b920 main (3, effffc94, effffca4, 5f3f4, 1, 0) + 2a0 00017b48 _start (0, 0, 0, 0, 0, 0) + 5c Ok, so what are we locked on? Parent process says: Current rlimit: 92 file descriptors 0: S_IFCHR mode:0620 dev:32,0 ino:162863 uid:100 gid:7 rdev:24,2 O_RDWR 1: S_IFCHR mode:0620 dev:32,0 ino:162863 uid:100 gid:7 rdev:24,2 O_RDWR 2: S_IFCHR mode:0620 dev:32,0 ino:162863 uid:100 gid:7 rdev:24,2 O_RDWR 4: 0xd000 mode:0444 dev:164,0 ino:8952 uid:0 gid:0 size:0 O_RDONLY close-on-exec 15: S_IFCHR mode:0000 dev:32,0 ino:26208 uid:0 gid:0 rdev:42,8491 O_RDWR 16: S_IFREG mode:0644 dev:162,1 ino:2924135 uid:0 gid:1 size:13387 O_WRONLY|O_APPEND 17: S_IFCHR mode:0000 dev:32,0 ino:6664 uid:0 gid:0 rdev:42,8911 O_RDWR 18: S_IFREG mode:0644 dev:162,1 ino:4842016 uid:0 gid:1 size:61789 O_WRONLY|O_APPEND 19: S_IFREG mode:0644 dev:162,1 ino:4842022 uid:0 gid:1 size:4551030 O_WRONLY|O_APPEND 20: S_IFREG mode:0644 dev:162,1 ino:4842019 uid:0 gid:1 size:29387 O_WRONLY|O_APPEND 21: S_IFREG mode:0644 dev:162,1 ino:4842021 uid:0 gid:102 size:0 O_WRONLY advisory write lock set by system 0x7FFF process 2135 Inode 4842021 is: 4842021 -rw-r--r-- 1 root devel 0 Feb 3 15:34 logs/.nfsCAC So, for some reason the accept queue mutex mechanism is no longer working. There are no lock manager error messages from the NetApp, but this server which has been running fine for months will suddenly no longer lock. Other servers sitting right next to it with the same configuration lock fine. Would be nice to determine why it stopped working, but the solution is to add the the following line to the Apache httpd.conf file: LockFile /tmp/accept.lock That will keep the lockfile on the local disk. This is probably not a bad idea for anybody running Apache off of a Netapp to do. It may save you some future headaches. -Rasmus

2 1

Apache/Netapp problem + solution
by rasmus＠bellglobal.com 04 Feb '98

04 Feb '98

Apache 1.2.5 on a Solaris 2.5.1 Sparc box connected to a Netapp 630 filer. Problem symptom: Web server established a connection, but would not respond to requests. Looking at the parent httpd process we see that it is in a normal "wait for something to do" loop: 0.000000 sigsuspend([] <unfinished ...> 0.415977 --- SIGALRM (Alarm Clock) --- 0.000214 <... sigsuspend resumed> ) = -1 EINTR (Interrupted system call) <0.415636> 0.000272 setcontext({uc_sigmask=[ALRM], ...}) = ? <0.000113> 0.000525 alarm(0) = 0 <0.000083> 0.000389 sigprocmask(SIG_UNBLOCK, [ALRM], NULL) = 0 <0.000087> 0.000475 sigaction(SIGALRM, {SIG_DFL}, NULL) = 0 <0.000086> 0.000464 waitid(P_ALL, 0, {si_signo=0, si_code=SI_USER, si_pid=0, si_uid=0, ...}, WNOHANG|WEXITED|WTRAPPED) = 0 <0.000091> 0.000499 alarm(0) = 0 <0.000084> 0.000349 sigaction(SIGALRM, {0xef5b8cfc, [], 0}, {SIG_DFL}) = 0 <0.000094> 0.000570 sigprocmask(SIG_BLOCK, [ALRM], []) = 0 <0.000090> 0.000577 alarm(1) = 0 <0.000086> 0.000351 sigsuspend([] <unfinished ...> 0.995342 --- SIGALRM (Alarm Clock) --- 0.000184 <... sigsuspend resumed> ) = -1 EINTR (Interrupted system call) <0.995330> 0.000248 setcontext({uc_sigmask=[ALRM], ...}) = ? <0.000107> 0.000517 alarm(0) = 0 <0.000083> Looking at the spawned child httpd's we see: 0.000000 fcntl(21, F_SETLKW, {type=F_WRLCK, whence=SEEK_SET, start=0, len=0} And they got there by: ef5b6fd8 fcntl (15, 7, 5cfcc) ef5b6fd8 _libc_fcntl (15, 7, 5cfcc, 0, 0, 0) + 8 ef78aba8 s_fcntl (15, 7, 5cfcc, ef611938, 34, effff944) + 164 00019138 accept_mutex_on (62d08, 1, 0, effffa78, effffa88, 2) + 18 0001a874 child_main (6c170, 5d000, 5d000, 5cc00, 5d000, 5b000) + 1f8 0001ad1c make_child (61880, 2, 61400, 11, 0, 5cfc9) + d4 0001b3a0 standalone_main (effffb78, 5cf48, 5f400, 5cc00, 75, 5f7b0) + 2a4 0001b920 main (3, effffc94, effffca4, 5f3f4, 1, 0) + 2a0 00017b48 _start (0, 0, 0, 0, 0, 0) + 5c Ok, so what are we locked on? Parent process says: Current rlimit: 92 file descriptors 0: S_IFCHR mode:0620 dev:32,0 ino:162863 uid:100 gid:7 rdev:24,2 O_RDWR 1: S_IFCHR mode:0620 dev:32,0 ino:162863 uid:100 gid:7 rdev:24,2 O_RDWR 2: S_IFCHR mode:0620 dev:32,0 ino:162863 uid:100 gid:7 rdev:24,2 O_RDWR 4: 0xd000 mode:0444 dev:164,0 ino:8952 uid:0 gid:0 size:0 O_RDONLY close-on-exec 15: S_IFCHR mode:0000 dev:32,0 ino:26208 uid:0 gid:0 rdev:42,8491 O_RDWR 16: S_IFREG mode:0644 dev:162,1 ino:2924135 uid:0 gid:1 size:13387 O_WRONLY|O_APPEND 17: S_IFCHR mode:0000 dev:32,0 ino:6664 uid:0 gid:0 rdev:42,8911 O_RDWR 18: S_IFREG mode:0644 dev:162,1 ino:4842016 uid:0 gid:1 size:61789 O_WRONLY|O_APPEND 19: S_IFREG mode:0644 dev:162,1 ino:4842022 uid:0 gid:1 size:4551030 O_WRONLY|O_APPEND 20: S_IFREG mode:0644 dev:162,1 ino:4842019 uid:0 gid:1 size:29387 O_WRONLY|O_APPEND 21: S_IFREG mode:0644 dev:162,1 ino:4842021 uid:0 gid:102 size:0 O_WRONLY advisory write lock set by system 0x7FFF process 2135 Inode 4842021 is: 4842021 -rw-r--r-- 1 root devel 0 Feb 3 15:34 logs/.nfsCAC So, for some reason the accept queue mutex mechanism is no longer working. There are no lock manager error messages from the NetApp, but this server which has been running fine for months will suddenly no longer lock. Other servers sitting right next to it with the same configuration lock fine. Would be nice to determine why it stopped working, but the solution is to add the the following line to the Apache httpd.conf file: LockFile /tmp/accept.lock That will keep the lockfile on the local disk. This is probably not a bad idea for anybody running Apache off of a Netapp to do. It may save you some future headaches. -Rasmus

1 0

strange cached information for CIFS
by Tom Yates 28 Jan '98

28 Jan '98

i have two F330s, both at 4.3R4, both running CIFS. apart from the number of shares on each one, and the machine names, the CIFS setups are supposed to be identical. one will authenticate through a PDC, and one won't. we have been through the old "create a domain account for the server" thing *many, many* times. it's not helping. when we try to "cifs setup" to use the PDC, we get the following joyous dialogue: Tue Jan 27 18:07:18 EST [rc]: Connection with DC BIGDADDY terminated Tue Jan 27 18:07:18 EST [rc]: CIFS - Filer retrying authentication with backup password. Tue Jan 27 18:07:28 EST [rc]: Connection with DC BIGDADDY terminated Tue Jan 27 18:07:28 EST [rc]: CIFS - Filer's security information differs from domain controller. and, in many cases, cifs will sit there trying to shut down, locking up the console, until i reboot the filer. it's been suggested to me that it needs to authenticate as part of the shutdown process. so, my questions are: 1) what flaming backup password? 2) what decides whether cifs starts up at boot time, absent a "cifs on" rc entry? 3) is it true that the delayed shutdown is caused by failure to authenticate? 4) has anyone else had trouble going from X to 4.3R4, esp. if X=4.0.1c{,D10}? and, if were allowed a slightly less serious wish list, any chance of an option such as options no.stupid.microsoft.authentication on to get the CIFS support to authenticate based on the user logged into the NT Workstation in question, a la NFS? or do i misunderstand terribly what's going on here? Tom Yates - Unix Chap - The Mathworks, Inc. - +1 (508) 647 7561 MAG#65061 DoD#0135 AMA#461546 1024/CFDFDE39 0C E7 46 60 BB 96 87 05 04 BD FB F8 BB 20 C1 8C

1 0

Faq(?) - Sendmail & Toasters getting along well?
by Don Glascock 23 Jan '98

23 Jan '98

Hi. We're evaluating/purchasing an F230, and are slowly getting all of our i's dotted and t's crossed in setting it up. In our environment, the sendmail.cf files on our Unix boxes hand everything over to the main file/mail server in our group, which is a DEC Alpha running Digital Unix. The workstations mount its "/usr/spool/mail" as "/usr/spool/mail" or "/usr/mail", as you would expect. With this configuration, we've had very few file locking problems, presuming that life is easier if the sendmail daemon that actually manipulates mail files runs on the same box on which the files sit. Now, we're debating about moving everyone's mail folders to the F230, which means that sendmail on our Alpha will be making regular use of NFS file locking. Has anyone tried this? How frequently (if ever) do you run into problems? Thanks & hope this isn't a FAQ... Don Glascock (ps, I'm new to the list -- is there an archive for this mailing list? I didn't see one on ftp.mathworks.com, and Majordomo says there are no "toasters"-related files. Thanks, again! :-) -- Donald S. Glascock Mailstop GU-1016 Special Purpose Processor Development Group Mayo Foundation Rochester, MN 55905 Glascock.Donald(a)Mayo.EDU +1.507.284.1850 +1.507.284.9171 (fax) http://www.mayo.edu/sppdg/sppdg_home_page.html "No matter where you go, there you are."

3 2

F630 with more than 52 disks ?
by Daniel Quinlan 23 Jan '98

23 Jan '98

I'm not planning on it, but I'd like to know if anyone out there has an F630 (or anything) with more than 52 9GB disks on it (~ 200GB). If you do, how long does a disk recontruction take? Our F630 with 52 9GB disks took 25 hours to reconstruct a failed disk (while it was being used, so we had to turn the knob down to 2 and 1 for most of the time because response was so slow). Dan

1 0

Solaris 2.6 drops NFS automounts
by Mike Mueller 23 Jan '98

23 Jan '98

We have heard reports that Solaris 2.6 NFS clients drop mounts of NetApp and other NFS servers but I have not been able to get a detailed explanation of the problem. Although it is apparently a Solaris 2.6 bug, I can't seem to find information about it from Sun and I thought that some other NetApp user might be familiar with the problem. We are planning a Solaris upgrade in the next few weeks and I'd like to feel confident that we aren't going to run into problems. Can someone explain what the specific bug is and whether there is a definate work-around or patch available? Is anyone successfully using Solaris 2.6 clients with a NetApp and if so are you using any special settings on either the client and/or the server? Thanks. Mike Mueller

2 1

Re: Anybody besides me having quotas go bonkers?
by Jan-Pieter Cornet 23 Jan '98

23 Jan '98

In article <199801202125.NAA26037(a)tooting.netapp.com>, Rajesh Sundaram <rajeshs(a)netapp.com> wrote: > attribution missing, but someone(a)cdsnet.net wrote: >> I just created some new users, and the filer (running 4.3R4D1) >> is reporting that they've used: >> >> user 506 9840 41943040 1292 10485760 >> user 1001 4294966368 41943040 22 10485760 >> user@web 1001 4294966408 25600 18 1000 >> user 1003 4294967260 41943040 1 10485760 >> user@web 1003 4294967260 25600 0 1000 >> user 1004 4294967260 41943040 1 10485760 >> user@web 1004 4294967260 25600 0 1000 >> >> which is flatly impossible, the users were just created and assigned. yet >> it thinks the disk space usage is all used up. We have been seeing the same thing last week. We fixed it by putting a (root-owner) file in the users' homedirectory, then chowning that file to the user, making their disk usage positive again. >Are default quotas being used here? If that's the case, this is likely to >be the bug with operations running as root that others in the list >have pointed out. > >Default quotas trigger the creation of quotas when a user first >uses an inode or a data block (by creating a file, for example). >This should happen even when the file is created as root and then >"chown"ed to the user. This currently fails to happen. > >Here is the description from the bug report which should soon get >added to the bugs online list on NOW. > >TITLE: >Derived quotas do not get created for operations >running as root, causing usage decrepancies. We upgraded to 4.3R4D2, which apparently fixed it. At least, I haven't seen "negative" disk usages in almost a day, whereas our previous version (4.3R4D1) would have had some by now. >> What's even weirder is that I can create directories all day, and no >> errors, only files cause problems. yet the dirs are owned by user 1001, >> etc. Toasters do not count directory blocks as disk usage, so directories can't be affected by this bug. -- #! ##### Jan-Pieter Cornet ##### <johnpc(a)xs4all.net> ##### perl ++$_;$!=$_+++$_;($:,$,,$/,$*)=$!=~/.(.)...(.)(.).(.)/;$!=$_+$_; ($(a),$\,$~)=$!=~/(.)(.).(.)/; $_="$,$/$:"; $@++; $~="$~$_";($_)= \$$=~/\((.)/;$|=++$_;$_++;$|++;$~="$~ $@$:";`$~$/$\$*$, $|>&$_`

1 0

4.3R4 and statd
by Michael Salmon 22 Jan '98

22 Jan '98

We have just installed 4.3R4 in a 520 with storage shelves 2 and statd has begun complaining about the shelves. Shelfchk has also stopped working so we are forced to conclude that the connection between the filer and the shelves has stopped working. Any comments or suggestions? /Michael -- /--------------------------------\ ___ | Michael Salmon | / \ | Ericsson Utveckling AB | / | | ÄS/UAB/Z/IPE | / _| | Box 1505 | / | | 125 25 Älvsjö | / / | Sweden | / / | | | | +--------------------------------+ | \ | Michael.Salmon(a)uab.ericsson.se | / *| | | / _/ | Tel: +46 8 727 30 74 | | | | FAX: +46 8 647 82 76 | | / # | GSM: +46 705 902 842 | | __// \--------------------------------/ |_/

1 0

Archived mail
by Dawn Hamel 22 Jan '98

22 Jan '98

Is the mail from this list archived anywhere. I believe there was some discussion about problems upgrading to Solaris 2.6.1 and I want to make sure a new Netapp customer is aware of any problems BEFORE they upgrade to the 2.6.1 release Thanks, Dawn

8 9

Jump to page:

2026

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