X/Open API on Linux
All, I'm working with software(AFS/OpenAFS) that uses the X/Open XBSA API to communicate with the TSM server. The manual Using the Application Programming Interface, Version 5.1 says that the X/Open API is only available for AIX, HP-UX and Solaris. Can someone (possibly Andy R.) comment on whether this functionality will be available for the Linux platform and if so what the time frame will be? Thanks, -Matt
Re: Exclude SYSTEM_OBJECT redux
If anything, I'd go with backupsystemobject yes/no, similar to backupregistry yes/no that we have today. Easy to put in a client option set, or in the options file on the client. I agree that this type of option would be preferable to my suggestion of exclude.systemobject. But is it possible with more recent versions of the server to put backupregistry in the client option set on the server? I run TSM server 4.1.0.0, and it is not possible to use that option in a client optioin set at this level. This must have been a feature they added in later versions. -Matt
Re: Idea for a TSM feature
The feature that I would like to see is the ability to create a special disk storage pool, that would only migrate inactive versions of backup objects to the next storage pool. This would keep all the active versions on disk storage. I'm sure everyone has certain nodes(systems) that are of more importance than others. Could be the CEO's desktop filesystem or a critical server. James, Have you considered colocation for the important servers and management desktop machines? I understand your point about not streaming the data as fast as a migration, but if implemented correctly these systems will each have their own set of tapes, and not require many tape mounts. Are you really measuring your restore in such small increments that 2 or 3 tape mounts is unacceptable? -Matt
Exclude SYSTEM_OBJECT redux
Folks, We are apparently in the minority due to the fact that we don't want to backup the SYSTEM_OBJECT on Windows 2000 on our clients. I have contacted Tivoli support and know that my only option currently is to use the DOMAIN statement in the dsm.opt file to control this. Therefore I come to the community and ask what (if any) requests have you made of the developers for design changes concerning this topic. I know some developers frequent this list, and if any of them care to comment on this, I would love to have your input. I plan on talking with Tivoli Marketing soon, asking for a design change to incorporate a feature that will allow me to exclude the SYSTEM_OBJECT; so maybe something like an option called EXCLUDE.SYSTEMOBJECT. It is my opinion that this should be an option that could be set from a server client option set, so that each client dsm.opt file doesn't need to be modified. This is a problem in my environment, as I have over 10,000 clients total, and it's difficult to make changes that require user action. When I spoke with Tivoli support about my problem, he mentioned that when designing the capability to backup the system files, they were required to take an 'all or nothing' approach. While this seems to be true, they make it quite difficult for the user to choose the 'nothing' route, much different from the philosophy TSM normally espouses. OK, thanks for listening. I look forward to your comments. -Matt
Re: volhist file extremely large
Would this also be applicable if all data was going directly to tape? and data is being archived for 3 and 8 years? Bill, I think what the consensus is here, is the tapes that are legitimate data tapes will not be deleted from the volhist file. Only DB backups, exports and volumes that have been rewritten to are going to be removed. Obviously then, the rewrite is when the volume is marked as STGNEW the most recent time in the file. -Matt /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
Re: Listserver question
Is it possible to see also your own postings to the listserver? Sometimes the posting doen't arrive, so now i must check it with an other account. Maybe you can use the web interface to the list at www.adsm.org to check your postings ? While we're on this subject...I frequently use the searching capabilities of this site, but notice that the hits are all listed twice. Is there anyone who is involved with this web site monitoring this list? If not I'll eventually go and post a note to the webmaster. -Matt
Re: Listserver question
Andy, Interesting indeed. I have done a little experimenting since your post. If I just search on Matthew and Bacchi I get many posts, with duplicates. Here is one particular selection, a single post that shows up twice. Notice how it has two different confidence levels for the same posting, wierd. 76% Re: Expire Logs C Program for Tivoli 4.2 and Informix ([EMAIL PROTECTED]) - Thu, 26 Jul 2001 20:08:43 GMT 60% Re: Expire Logs C Program for Tivoli 4.2 and Informix ([EMAIL PROTECTED]) - Thu, 26 Jul 2001 20:08:43 GMT But if I search for Matthew AND Bacchi I get fewer responses, and none are duplicates. It appears to me that over the years I have been using this search without the keyword AND in caps. For whatever reason, the search engine interprets this differently than I expected. Does this appear to be normal behavior to you? The strange part is though, why would the lower case word and cause the engine to spit out the same post twice? If the maintainer does follow this discussion, maybe they would like to comment. -Matt /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
volhist file extremely large
Folks, I have seen some posts concerning this in the past, but I don't have a good feeling about my options. I have a volumeshistory file that is around 5 megs, and entries that go back to 1997. What I want to know is, how do I know for sure that I can delete a particular entry? Kelly mentioned that he does a delete volhist type=all todate=today-30 but that seems a little extreme to me. I believe that I have volumes that have data on them for much longer than 30 days. I am pretty sure they haven't had data on them for 4 years, but I'll bet it is closer to 6 months. What do you guys do to verify that the volumehistory entry can be deleted? I suppose I can write a script, but wanted some input first. Thanks in advance. -Matt /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
disabling TCPIP commmethod
OK folks, this one has bothered me for years. I want to disable the TSM server completely while I do some work. Rather than using the disable sessions command, I want to in effect disable TCPIP connections altogether. I have attempted commenting out COMMmethod TCPIP in the dsmserv.opt file, but it appears that this is the default, so it has no affect. Besides the obvious answer of disconnecting the network cable, how can I force the TSM server to ignore every TCPIP request. When sessions are disabled, the activity log shows connection started, than a message indicating that user access is disabled. I want to avoid this. Thanks in advance. -Matt /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
ANS1809E error on AIX client
Folks, I have been seeing what appear to be networking problems on an AIX client over the past 2 weeks. An incremental normally runs from 06:00 AM until 19:00 PM, but recently the process has not finished by the next morning at 06:00. We get errors in our logs, and redirected from standard out. Here are some of them: In dsmerror.log: 07/28/01 12:50:36 TcpFlush: Error 32 sending data on Tcp/Ip socket 4. 07/28/01 12:50:36 sessSendVerb: Error sending Verb, rc: -50 07/28/01 12:50:36 TcpFlush: Error 32 sending data on Tcp/Ip socket 4. 07/28/01 12:50:36 ANS1809E Session is lost; initializing session reopen procedure. 07/28/01 12:50:36 ANS1809E Session is lost; initializing session reopen procedure. 07/28/01 12:50:56 ANS1810E ADSM session has been reestablished. From standard out: ANS1809E Session is lost; initializing session reopen procedure. A Reconnection attempt will be made in 00:00:14 A Reconnection attempt will be made in 00:00:13 A Reconnection attempt will be made in 00:00:12 ... A Reconnection attempt will be made in 00:00:00 ... successful This error happens over and over in our logs, until the process has to be killed the next day. Has anyone seen these types of messages, and if so, what was your fix? Also, it should be noted that the client is on AIX 4.3.3, running the ADSM 3.1.0.8 client. The idle timeout value is 120 minutes and the commtimeout is 180 minutes. The client is on 10 Meg Half duplex ethernet, with an MTU of 1500, and the server is on FDDI with a MTU of 1492. I have used ping to figure out if there is any packet loss, and find none with normal size packets and packets of 4000 bytes in size. Thanks in advance for any help that you can give. -Matt /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
Re: ANS1809E error on AIX client
Good point Richard. I do see the session being terminated on the server, here is the log entry: 07/28/01 12:50:37 ANR0480W Session 56995 for node MDBACK0.BTV.IBM.COM (AIX) terminated - connection with client severed. Obviously because this is from Saturday, I don't know for sure if the storage pool is full(it is a disk pool), but I doubt it, this is a 215 Gig pool, and the hi migration pct is set to 20%. Thanks again, -Matt Matt - Big missing piece in your posting: what's in the server Activity Log at that time? Most of the time, it is the server killing off sessions for some reason, like tape errors or storage pool full. Richard /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
Re: AIX TSM scheduler problem
Be aware that inittab runs things under sh, not ksh. See man page. Richard, Not to split hairs here, and dangerously close to being off topic, but isn't sh linked to ksh? From the man page: man sh: Syntax Refer to the syntax of the ksh command. The /usr/bin/sh file is linked to the Korn shell. Description The standard configuration of the operating system links the /usr/bin/sh path to the Korn shell. ls -l : -r-xr-xr-x 4 bin bin 240390 Dec 02 1999 /usr/bin/sh -r-xr-xr-x 4 bin bin 240390 Dec 02 1999 /usr/bin/ksh that is on a AIX 4.3.3 system at least. Maybe if your system is configured differently you would have a different sh binary. -Matt /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
Re: export multiple nodes, but only import one?
Jack, This creates two questions. If I export a dozen nodes to a tape, and then later someone needs data from node XYZ, can I stick the tape in and IMPORT NODE XYZ DEVCLASS=8MM_CLASS and it will ignore the other nodes? In my experience this works exactly as you describe. You can specify what node will be imported even though multiple nodes have been exported to one tape. EXPORT NODE A,B,CFILEDATA=ALL 8MM_CLASS SCRATCH=YES EXPORT NODE D FILESPACE=D1,D2,D3 FILEDATA=ALL 8MM_CLASS SCRATCH=YES Will this back it all up to the same tape and then allow me to put in the tape and import one specific node or filespace? This scenario on the other hand is more of a stretch. I don't know for sure if this will work as you expect. The main reason for this, is that the two export commands might want to start from the beginning of the tape. If you are interested in utilizing the tapes efficiently, you may want to approach this problem from a slightly different angle. If you export your node data to a FILE device class, then tar all that data to your tapes(spanning tapes if necessary), you get the greatest efficiency and can later untar just what you desire before importing to TSM. The only problem you may have is that your disk capacity is not enough to handle the exports from the TSM server. Good luck. -Matt /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
Re: Expire Logs C Program for Tivoli 4.2 and Informix
Fred, From what I have read and been told by IBM support, you will want to use the TSM 3.7 API no matter what version of the other TSM client code you use. Apparently the API for 4.1 and above has major problems, this will be fixed, but for now just use the 3.7 API. -Matt /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
Re: ADSM/HSM migration candidates list
Matthew - Is the file system occupied by small files, under 4KB in size? No, it isn't. In fact it is occupied by a few quite large files, between 100-500 Meg. There are some lock files in the filesystem, which are around 4KB, but the majority of the files are larger. Richard, maybe you or someone else can answer this question for me. The way I understand the product to work, it will always generate a migration candidates list based on the Age and Size factor settings, even if your thresholds are set so that migration will not occur. The candidates list is just a list of files and their score, and doesn't indicate that migration is eminant. I have configured this product on quite a few other machines in the past, this one just seems to be giving me trouble. Thanks again, -Matt /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
Re: help to certification
Do you know if there are some web sites where it's possible to try = tests for TSM certification ? Yep, try this Tivoli site: http://www.tivoli.com/services/certification/exams/699exam.html You'll see the Sample Exam in PDF format. -Matt /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
ADSM/HSM migration candidates list
Folks,
I can't seem to get HSM to generate a candidates list and begin migrating
files on an AIX 4.3.2 system. The version of adsm.client.aix42.hsm is
3.1.20.8, one of the most recent ADSM packages. Anyway, the high migration
threshold is set below the percent used for the filesystem, Space Management
is active for the filesystem and we have backed the files up using the normal
ADSM b/a client(even though the MIGREQUIRESBACKUP parameter is set to NO.)
Here are some commands and their output:
{ root@client } # dsmreconcile -c /b32tmd/tmd3
Reconciling '/b32tmd/tmd3' file system:
Querying the ADSM server for a list of backed up files...
Received 11616 entries
Now traversing the file system...
Writing the new migration candidates list...
Note: unable to find any candidates in the file system.
ANS9250I File system '/b32tmd/tmd3' reconciliation completed.
{ root@client } # dsmmigquery -c /b32tmd/tmd3
Candidates for migration in file system /b32tmd/tmd3:
no file exists in the migration candidates list
{ root@client } # dsmmigfs query /b32tmd/tmd3
File System HighLow Premig Age SizeQuota StubServer
NameThrshld Thrshld Percent Factor Factor SizeName
/b32tmd/tmd385 65 5 1 0 19000 4095ADSM0
{ root@client } # dsmdf /b32tmd/tmd3
FSM FS Mgrtd Pmgrtd Mgrtd Pmgrtd Unused Free
Filesystem State KB KB Files Files Inodes KB
/b32tmd/tmd3 a 0 0 0 0 2356854 867700
{ root@g54srv1 } # df -k /b32tmd/tmd3
Filesystem1024-blocks Free %UsedIused %Iused Mounted on
/b32tmd/tmd3 9502720867704 91%18825 1% /b32tmd/tmd3
Anybody have any ideas? I have spent quite a few hours now playing with this,
making changes and attempting to generate a candidates list, all to no avail.
Maybe I need another set of eyes. Thanks in advance.
-Matt
/**
**Matt Bacchi [EMAIL PROTECTED]
**IBM Global ServicesSDC Northeast
**F6TG; MD Filesystems/Internet (802) 769-4072
**ADSM AFS/DFS Backup (tie) 446-4072
**/
TSM icon in gif or jpg format
Folks, I know this sounds silly, but I'm updating my internal web page for TSM, and can't find a TSM logo/icon anywhere. I do see the Tivoli logo, but nothing that is more specific to TSM itself. When ADSM was king, there was the Safeman icon that we used, but now I'm interested in moving into the 21st century. Anyone know where I can get one of these? Thanks, -Matt /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
Rebinding archive to management class
Folks, I know its not possible to rebind previously archived files to a new management class, but I want to try it anyway. My customer has been using the default management class when archiving in the past, but wants to increase the retention period from 180 to 730 days. I can't do this for the default management class, so I have decided to create another management class with a 730 day retention period. So, his newly archived files will be bound to this second management class, but his old ones will still be bound to the default class. To 'coerce' ADSM into rebinding his previously archived files I was thinking of creating a new policy domain, and associating the node with this domain. In this case, all archive files will become rebound to the management class with the same name as in the old policy domain, or in the case that a management class of the same name doesn't exist, they will use the grace retention period. Am I on the right track here? Can anyone recommend a better way to do this? Thanks in advance, -Matt /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
Re: db2 backups not expiring
Peter, Another thing to think about is how often your database is backed up fully, versus the logs being archived. We only back up the dbs fully once per week, and logs are applied to get to a particular point in time. Because of this, we use : vere = 2 verd = 0 rete = 0 reto = 0 Also, this forces inactive copies to be deleted, so we don't keep 50 Gig databases around that don't need to be. -Matt Does anybody know why wouldn't db2 backups expire, when, in theory, they should be? We have an AIX-SP environment, with TSM 4.1, server and clients. DB2 data is backed up via the TSM-API. The copygroup criteria for this data is: versions exists=1; versions deleted=1; retain extra versions=1, retain only version=3. DB2 backups have been occurring weekly since mid-November 2000, yet none of the DB2 backup images have ever expired. We run inventory expiration daily. Any ideas? /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
TSM 4.1 API broken?
Folks, I thought I read somewhere that the TSM 4.1 version of the API was not working well, and it is recommended to use the 3.7 version. Is this correct or am I going crazy. I have tried to search through the archives on adsm.org, but get poor results when using anything with the word "api". If this is not correct, what is the error message below coming from: 03/29/01 02:36:39 ANS1036S Invalid option 'DSMSRVR' found in options file 'OPTIONS STRING' at line number : 1 Invalid entry : '-DSMSRVR=adsmsrv1.sg.ibm.com' This is when trying to start butc to backup AFS to a TSM server. Thanks, -Matt /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
Re: FIVE questions for TSM 4.1.2
George, I'm not really sure what the difference is, I'm guessing that the SSA controller needs to support Fast-Write Cache, and I believe it's a chip on the controller with memory(cache) for doing this. For example if I run the following command: # lscfg -vl ssa0 DEVICELOCATION DESCRIPTION ssa0 10-68 IBM SSA 160 SerialRAID Adapter (14109100) Part Number. 27H1204 FRU Number.. 34L5388 Serial Number...S0157201 EC LevelE28728 ManufacturerIBM053 ROS Level and ID8300 Loadable Microcode Level05 Device Driver Level.00 Displayable Message.SSA-ADAPTER Device Specific.(Z0)SDRAM=128 Device Specific.(Z1)CACHE=32 Device Specific.(Z2)UID=08005AEBED5A Device Specific.(YL)P2-I3/Q1 You see it's a model 160 adapter, and in the Z0 and Z1 fields it has the amount of SDRAM and CACHE on board the controller. Check yours out and see what it has. -Matt Matt, I have an even older 7133 D40 attached to a server and rather than have Fast Write enabled (slightly different syntax between machines), it has FL_UNAVAIL (fast load unavailable). Any idea what this means??? Thanks- /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
Re: AFS 3.6 butc deletedump
Is there a synchronization utility to force the synchonization of the TSM data with the AFS backup database? Kent, This is a good question, but I don't think there is. I have used BUTA for 1+ years now, and there is a sync option with the delbuta utility. When doing some testing with BUTC in the last six months I noticed there wasn't a similar function. I'm currently not using BUTC because the 'backup diskrestore' function doesn't work with BUTC on afs36, but this problem could be a show stopper as well. Anyway, I assume you have already read the Release Notes (http://www.transarc.com/Library/documentation/afs/3.6/unix/en_US/HTML/RelNotes/aurns002.htm) which describe the new options when using the 'backup deletedump' command, such as -dbonly and -force. You might also try writing you own syncronization script, which would consult the AFS backup database, then do ADSM "delete filespace" commands for each dump that isn't in AFS anymore. Or as a last resort, maybe you want to open a problem with support on this. Good luck, and let us know what you end up doing, I'd be interested. -Matt /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
Re: TSM training?
I would encourage you to look into going to the SSSI advanced class. I took it in October and was impressed. They covered new features of TSM, the Tivoli Data Protection clients and went into sizing of TSM servers extensively. I learned quite a few new things, and I've been doing ADSM for 3+ years. -Matt www.storsol.com We offer two courses: Level 1 Beginner and Level 2 Advanced. Taught by folks that implement TSM daily. The Advanced class covers SQL, Exchange and Oracle connect agents, DRM, Server to Server among other things. Taught in Colorado during four days. Kelly J. Lipp /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
Re: Delete of one volume in VOLHIST
Rene, You can use the command "delete volhist type=export" if you have only one export volume. Try first doing a "query volhist type=export". Hope this helps. -Matt how can I delete only one volume in the VOLHIST ? This is an Export volume, created 2 months ago. I cannot use the "dsmc del volh todate=today-60" for instance as I have to keep older export volumes ! /** **Matt Bacchi [EMAIL PROTECTED] **IBM Global ServicesSDC Northeast **F6TG; MD Filesystems/Internet (802) 769-4072 **ADSM AFS/DFS Backup (tie) 446-4072 **/
polling vs. prompted scheduling
I'm confused. I know what the two different scheduling methods are, but I don't think my server does. I have the server set to do client polling, but I still get error messages in the activity log that look like this: 01/11/01 15:03:39 ANR8214E Session open with 9.66.95.61 failed due to connection refusal. 01/11/01 15:03:39 ANR2716E Schedule prompter was not able to contact client JOHNPAULPC.BTV.IBM.COM using type 1 (9.66.95.61 1501). What does this mean? For all you who think I'm crazy, here's the output from the "q st" command: Scheduling Modes: Polling (well, it's amended) The strange thing is that the server has been set to client polling for years, so it's not a recent change. TIA -Matt Bacchi
HSM age factor
Hello all, We are attempting to understand the age factor parameter of HSM a little better. I know that the age factor is used to prioritize the files eligible for migration. Is there a way to make HSM look at the last access time of a file before it marks it as an eligible candidate to migrate? For example, we wish to keep files on the local filesystem for 90 days before having them migrate. Can this be done with just the age factor? My understanding of the age factor parameter says that this is not possible. Therefore I'm thinking I need to employ a script that will modify the EXCLUDE list until a file hasn't been used for 90 days. Then when I remove it from the exclude list, it should be migrated as an eligible file. Does anyone have any insight or opinions on this? Thanks, -Matt Bacchi *-* Matt Bacchi [EMAIL PROTECTED] IBM Global Services [EMAIL PROTECTED] Dept. F6TG, MD Filesystems/Internet (802) 769-4072 ADSM AFS/DFS Backup(tie) 446-4072
Upgradedb command
Hi, When doing an dsmserv upgradedb command, the server continues to run after completion. Is it possible to add a flag that will shut down the server once it has finished upgrading the database? We are attempting to automate some of the server upgrade process, and this one command is halting our progress 8^) Thanks, -Matt *-* Matt Bacchi [EMAIL PROTECTED] IBM Global Services [EMAIL PROTECTED] Dept. F6TG, MD Filesystems/Internet (802) 769-4072 ADSM AFS/DFS Backup(tie) 446-4072
Archive large list of files
Hello, We have a process that archives a large set of files. We are running into a problem now because the number of files is apparently too great for the command interpretor (shell) to accept as individual arguments. We use this particular method so that we can archive all these files with the same description, and using the same timestamp. Is anyone aware of a method of using an input file to list all the files required to be archived. If not, can we fool the shell by putting double quotes around all the files, essentially using this as one argument only? Any help would be appreciated. -Matt
