------------------------------------------------------------------------ NOV-MSC3.DOC -- 19980325 -- Email thread on miscellaneous NetWare topics ------------------------------------------------------------------------ Feel free to add or edit this document and then email it back to faq@jelyon.com Date: Thu, 8 Jan 1998 15:42:20 -0500 From: Andrew Wyman Subject: Re: OSI model questions An easy way to remember the OSI layers: All [Application] People [Presentation] Seem [Session] To [Transport] Need [Network] Data [Data Link] Processing [Physical] --------- Date: Thu, 8 Jan 1998 14:03:39 -0700 From: Joe Doupnik Subject: Re: OSI model questions >I'm studying for the Networking Technologies exam and have a couple of >questions. > >I'm also taking the Microsoft Network Essentials course. Novell's Net >Tech course says that gateways operate at the network layer of the OSI >model. Microsoft says that gateways operate the application layer. >I was thinking that because a gateway is a fancy translator perhaps >it really belongs at the presentation layer. > >Which is correct? More to the point, which answer is the correct >one for the tests? Most importantly all of the above are wrong. "Gateway" is a very generic term and protocol dependent for its meaning. Next, essentially no one writes code to the OSI model; it's long dead and buried except in say parts of DECnet that are ignored and in classes that don't know what they are doing. Routers operate with protocol headers and up, and also can behave as real Bridges (MAC headers and down). Protocol headers are one level up from MAC headers. That puts them near the network layer. The bothersome thing here is the temptation to memorize answers and reguritate same. Some tests are designed to reinforce this habit. Joe D. --------- Date: Thu, 8 Jan 1998 16:55:39 -0500 From: "Brien K. Meehan" Subject: Re[2]: OSI model questions >It's all the sublayers (topics and subtopics) within each layer. It's a >lot of material to memorize. Then you throw in the protocols and how >they match up to the OSI model...Whew! There's nothing to practice or >do, it's all strictly memorization...yuk! > >Any other hints/tricks/ideas? No, it's not strictly memorization. It might serve you well to gain some understanding. One good way would be to write a couple NLM's, and a couple applications that use multiple file handles, password authentication, direct IPX dialog, etc. - in assembler. That would give you a real good feel for what functions happen at which layer. ...but that's probably not convenient. Well, then, try to read what you can. If you have Lanalyzer, or some way of capturing packets, you could try capturing something simple, like a COPY command, and then an NCOPY command. Look at all the packets. You can see evidence of many layers there - Network, Transport, and Session should be clear. It could make it easier to understand what happened at the Physical, Data Link, Presentation, and Application layers to get that file from here to there. If you have a CLEAR understanding of how a file goes through ALL layers when you do a DOS COPY command, you should be okay. And don't skimp on understanding the Presentation and Application layers that exist on the server! You will hear that this model is purely academic, and otherwise not useful. Don't believe it. At least not fully. If you're going to be a CNE, you're probably going to be fixing things that used to work but have stopped. If you have a good understanding of HOW things work, it will help you understand WHY things don't. I got into a "discussion" on CNENET about how something dumb - I think it was about how the Netware OS makes an erased copy of a file before you save it, so that your old copy is still there, salvagable. This doesn't happen. That is, Netware doesn't do this, the workstation does this, and then only if the application tells it to. Most of the CNE's in that discussion didn't have a clear understanding of what functions occurred on what machine at what level, and I think it's because they memorized that OSI model stuff, and forgot it as soon as they got their "paper." Don't be like that. With a good understanding of the theory, in that example, you would KNOW why you can't salvage Word files, but you can if you EDIT files - instead of wondering where your files are. --------- Date: Fri, 9 Jan 1998 01:39:33 -0500 From: "John J. Tulko, Jr." Subject: Re: OSI model questions >>I'm also taking the Microsoft Network Essentials course. Novell's Net >>Tech course says that gateways operate at the network layer of the OSI >>model. Microsoft says that gateways operate the application layer. >>I was thinking that because a gateway is a fancy translator perhaps it >>really belongs at the presentation layer. Which is correct? More to the >>point, which answer is the correct one for the tests? > > Most importantly all of the above are wrong. "Gateway" is a very >generic term and protocol dependent for its meaning. Next, essentially no >one writes code to the OSI model; it's long dead and buried except in say >parts of DECnet that are ignored and in classes that don't know what they >are doing. > Routers operate with protocol headers and up, and also can behave >as real Bridges (MAC headers and down). Protocol headers are one level up >from MAC headers. That puts them near the network layer. > The bothersome thing here is the temptation to memorize answers >and reguritate same. Some tests are designed to reinforce this habit. > Joe D. While I do agree that the term gateway is generic, everything that I have learned states that gateways can work on the network layer, the transport layer, or the session layer. On a different note, I remember writing a back to someone that wanted to know if anyone had a quickie quiz/test for testing candiates for new positions, before they are hired. One of my questions were, "At what layer of the OSI model do routers work at?" Joe D replied, OSI is as dead as a door knob. I wrote back where did you get that information. While it may or may be not be true that no one actual writes code base on OSI, I still think you mean the ISO protocol is dead. They are two different things. As an example, TCP/IP, Novell's IPX, and ISO are all based now on the OSI model. Hit Network General's Sniffer HomePage, which is now part of McAfee, and you can request a nice poster of how all the protocols tie into each other. John "Still don't think OSI is dead" Tulko, CNE --------- Date: Fri, 9 Jan 1998 08:26:53 -0700 From: Joe Doupnik Subject: Re: ODI model questions >While I do agree that the term gateway is generic, everything that I >have learned states that gateways can work on the network layer, the >transport layer, or the session layer. > >On a different note, I remember writing a back to someone that wanted to >know if anyone had a quickie quiz/test for testing candiates for new >positions, before they are hired. One of my questions were, > >"At what layer of the OSI model do routers work at?" > >Joe D replied, OSI is as dead as a door knob. I wrote back where did >you get that information. While it may or may be not be true that no >one actual writes code base on OSI, I still think you mean the ISO >protocol is dead. They are two different things. As an example, >TCP/IP, Novell's IPX, and ISO are all based now on the OSI model. >Hit Network General's Sniffer HomePage, which is now part of McAffe, >and you can request a nice poster of how all the protocols tie into >each other. > > John "Still don't think OSI is dead" Tulko, CNE ------- I stand on what I said previously. TCP/IP and IPX are definitely not based on OSI work. TCP/IP preceeds that work. ISO and OSI are different terms, and convention has put OSI as the cover term for both. While it is true that OSI code is extremely difficult to locate and no new development has been observed does suggest to me and many others that the effort is dead. The protocols don't tie into each other, alas. Joe D. --------- Date: Fri, 9 Jan 1998 10:23:00 -0700 From: Hansang Bae Subject: Re: OSI model questions [snip: Def of GATEWAY is diff btn NT and NW test material] >really belongs at the presentation layer. Which is correct? More to the >point, which answer is the correct one for the tests? Whatever the test study material says. Unless you are prepared to argue the accuracies (or is it inaccuracies?) of the test. >Does anyone have any "tricks" to memorizing all the sublayers and >subtopics of the OSI model and how different protocols map to it? >I have learned the 7 layers themselves (PDNTSPA, the easy part), >it's all the topics and protocols associated with the different >layers that confuse me. Too much theory, nothing "hands on"...this >test is hard. But it's the theory part that can help you out in the end. Too many so called network engineers are really just network technicians. Too many people don't have the detailed understanding of the protocols, and how they work. If you know the details, it can help you in your troubleshooting. While JoeD is correct in saying that OSI is dead and buried, it is still a good learning tool. Like pascal, LOGO, etc. You learn to appreciate who is supposed to do what and how the big picture breaks down. The entire exercise can be done with the TCP/IP suite, but OSI is a bit more generic, methinks. Just a learning tool with no real world-ROI kinda-thingy! What's the best way to think about it? Study and understand it. It's kinda like math, if you know how to derive the formula, then you don't ned to memorize those pesky formulas. But as in math, deeper understanding only comes with later courses. Kinda like not truly understanding Calculus until you take Linear Algebra! oop ack! ------------------------------ Date: Fri, 9 Jan 1998 10:53:32 +0000 From: Randy Richardson Subject: Re: name space >I'm running NW 4.1 with a few ip services to integrate out NT Clients. >Which name space is the right or better choice to support long file >names: OS2 or NFS? > >At the moment I'm using OS2 name space, but the latest version of >vrepair always claims something about V_LONG. > >I know that OS2 name support has moved to LONG.NAM, is this useable >for 4.1 also? Or is the latest vrepair not for use on NW 4.1? > >What are the differences between OS2.NAM and NFS.NAM concerning the >usage on NT. My planning is to change all clients from dos/win3.11 >to NT, so don't take care of WIN95 and any other type of client. In NetWare 4.10 and older, use OS2.NAM. In IntranetWare 4.11 and newer, use LONG.NAM. VRepair uses V_OS2 to support OS2.NAM, and uses V_LONG to support LONG.NAM. ------------------------------ Date: Fri, 23 Jan 1998 23:21:16 -0700 From: Joe Doupnik Subject: Re: What is SPX/SPXS? >I find all manner of information on implementation, but nothing to >indicate what it is, what it does, or whether it's necessary on a given >server. > >The reason for my concern is that we don't have it loaded; but upgrades >are mentioned as a possible solution for a series of abends we've >suffered the last couple of days, involving STREAMS -Q and error code >00000000. I do not, however, want to start indiscriminantly loading >modules we don't employ. ---------- IPX deals with datagrams, packets sent with no historical ties. The TCP/IP equivalent is real IP datagrams, or even UDP over IP. SPX is the session oriented method where history and sequence numbers and flow control and retries and ACKs and all that good stuff become available. The analog to SPX is TCP over IP. That's assuming one knows about UDP and TCP and IP, of course. Streams is a way of organizing a protocol stack. No necessary relationship to SPX/IPX. One uses SPX most often with tape backups where flow control and assured error free delivery are paramount. As usual, first backup your machine thoroughly, then apply the issued patches and updates. Afterward folks may be willing to help sort out troubles. Joe D. ------------------------------ Date: Sat, 24 Jan 1998 15:25:24 -0000 From: George Abbot School Subject: Resizing a RAID 5 Netware Partition: Thanks & Howto (x3) Thanks to all those who replied to my query. Needing to resize on five new and identical servers gave me the chance to try all suggestions. Here's what happened. Problem: A RAID array 5x4Gb disks had been configured 500Mb DOS, 8Gb Netware, 8Gb free space to be set up as a second Netware partition. This "would allow us to get over the 8 volume limit". Since the RAID was configured as one logical partition it presented to Netware as one drive and hence a second partition could not be created, nor the original resized. PREcautions: SBackup every server; install client on DOS partition every server and use it to copy SYSTEM and LOGIN from SYS: of next server in the ring onto the DOS partition into a SYSBACK directory; . Solution 0: (didn't work) Use the RAID configuration utility to redefine the original logical partition onto a smaller number of physical disks and create a second logical partition. On re-initialisation all data was lost and we started from scratch with a bare machine. Solution 1: Load TSAetc and SBACKUP and prepare to do a restore of SYS: volume; load VRepair just in case; add search path for c:\sysback; load INSTALL; used INSTALL to save NDS to DOS partiton; dismount volumes; delete volumes; delete Netware partition; add Netware partition; recreate volumes; tried to restore SYS: but could not authenticate and could proceed no further. Brought down server; brought up server "by hand" using the c:\sysback files; used INSTALL to install basic system files to SYS: and bring back NDS from DOS partition; used SBACKUP to restore SYS:; brought down and restarted server; backed up volumes were restored; VRepair reported no problems. Solution 2: Load TSAetc and SBACKUP and prepare to do a restore of SYS: volume, including authenticating to source and target; load VRepair just in case; add search path for c:\sysback; load INSTALL; used INSTALL to save NDS to DOS partiton; dismount volumes; delete volumes; delete Netware partition; add Netware partition; recreate volumes; restore SYS: without session files; at end of restore server Abended gracefully but restarted happily (save I needed to reinstall licences); backed up volumes were restored; VRepair reported no problems. Solution 3: Used an evaluation copy of SNAPBACK from www.cpd.com. This includes a "DOS" utility called NETSIZER which, they tell me, is not available separately. Running Netsizer displays DOS and Netware partition size as an FDISK partition together with the remaining free space and allows the latter to be allocated to the former. On restarting server, INSTALL sees the free space as part of the Netware partition and allowed it to be assigned to new and existing volumes. VRepair reported no problems. A real cinch. All servers have been happy for a week. Only solution 3 required no NWAdmin corrections to the NDS. ------------------------------ Date: Sun, 8 Feb 1998 09:51:05 +0100 From: Steinar Kleven Subject: ContextLess login available for free ContextLess login for Win95 and NT is available for free from my page: http://www.ahs.hist.no/distr/NWTools/ ------------------------------ Date: Thu, 12 Feb 1998 23:15:24 +1300 From: "Baird, John" Subject: Re: Migrating 3.12! to new server results in bad bindery >While we're on the topic: can anyone pls explain why when we run >BINDFIX we get a list of "Discarding object node with Invalid ID" >messages, and then if we run it again straight away, we get exactly >the same number of messages? BINDFIX is discarding objects in the bindery with names containing characters not normally allowed in an object name. These are dynamic objects created in response to receipt of SAPs. Dynamic objects are not subject to the same name restrictions as 'normal' objects created via SYSCON or PCONSOLE, but BINDFIX overlooks that. The objects are recreated next time SAPS are received, ususally no more than 60 secs later, which explains why the same objects are removed each time BINDFIX is run. ------------------------------ Date: Wed, 18 Feb 1998 12:03:33 -0600 From: Carl Milloshewski Subject: Re: Low Budget NetWare Testbed Not to knock Partition Magic, but my opinion for the best boot manager is System Commander by v-communications. http://www.v-com.com I run Linux, Dos/Win, Win95, NT 3.51 Server, NT 4.0 Server, NT 4.0 workstation, Novell 3.12, and Novell 4.11 on one computer with two disk drives. One 2 Gig SCSi and a 540 IDE. By far the easiest way to multiboot! ------------------------------ Date: Sat, 28 Feb 1998 15:03:32 +1300 From: "Baird, John" Subject: Re: Rconsole ... Yes , Slist ... No >NW411 server. > >I can load the lan drivers, bind ipx, load remote and rspx and >rconsole in w/o problems. > >What can't I do? I can't see the server using slist. I can't restore >a deleted SYS volume w/Arcserve until; I can see it. Slist uses bindery based calls to query the server your workstation is connected to for known objects of type 4 (file server). These are dynamic objects created in response to receipt of SAPS from those servers, or from routers. So, either your problem server is not sending out SAPs, or the server you are connected to is not hearing them which suggests a frame type mismatch. ------------------------------ Date: Sat, 28 Feb 1998 21:48:16 +0800 From: Cordelio Sumbillo Subject: Re: Monitor in 4.10 vs. 4.11 >We have just recently upgraded 2 of our Novell 4.10 servers to 4.11. >I have noticed now that in Monitor under the connections users are >not alphabetized like they were in 4.10. Just wondered if we missed >something during the upgrade or there is possibly a patch for this. Actually, you did not missed anything. You can sort alphabetically by pressing F3, sort options as shown in your monitor console. ------------------------------ Date: Sun, 1 Mar 1998 16:56:42 -0700 From: Joe Doupnik Subject: List's FAQ, now web searchable Just a small note to say I added a search engine to netlab1.usu.edu so that the list's FAQ can be searched for key words and phrases by web browsers which support Forms. The "home page" offers four styles of seaching, in order of effectiveness from left to right. It's not perfect but it works. I hope this will encourage folks to check the FAQ before asking what might be a well discussed item. Joe D. P.S. Suggestions on the engine, pitched at simple enough level that even I can understand, will be appreciated. This is not my favorite activity. ------------------------------ Date: Wed, 4 Mar 1998 14:59:40 +0200 From: Mike Glassman - Admin Subject: Re: Netware 4.11 SFT3 Problem - HELP!!! >I am in desperate need of some help RE: a NetWare SFTIII setup: > >Problem: The secondary file server completes a mirror, then when >files are accessed they appear corrupted. Files run or accessed >from the server are corrupted. Also, NLM's loaded at the server >are also corrupted. Basically all files on the mirrored partition >are not in any usable form. At this point if the secondary file >server is downed, and everything runs beautifully again. The >secondary server crashes constantly most of the >time as well (but not during the re-mirror process). > >What am I doing wrong? > >Background: Both servers are completely identical. Both have the >following >configuration: >Hardware: >-Pentium II - 233 MHz CPU >-96 MB SDRAM >-1x 6.4GB Quantum Fireball SE IDE Hard Drive (Installed on Primary IDE >as Master) >-1x Panasonic 24x IDE CD-ROM (Installed on Secondary IDE as Master) >-3x Accton 1207 NIC's (DEC 21041 chip based 10/100 Ethernet cards) (1 >used as MSL link, 1 used for 10Mb segment, 1 used for 100Mb segment) > >Configuration: >-Both servers have Identical NCF files (with the exception of the >first IO engine which also loads APC's powerchute software). >-Both servers are fully patched with the latest service patch from >Novell. >-There is an 80 MB DOS Partition on each HDD (identical). >-There is only one NetWare volume on the HDDs using the remaining >space (identical). >-Long filename support has been added to volume SYS. >-All NIC drivers are loaded in the iostartups, with the MSL driver >loading last. All protocols are bound in the IOauto. >-All HAMs are loaded in iostartups, CDMs in MSstartup. > >-ALL Hardware has been through thorough diagnostics / burnin and all >are >functioning correctly. NIC cards included. >-There is NO ACTIVITY on the server(s) while remirroring is >established, and >all logins are disabled. > >Why is my mirror drive not working? Why is it corrupting all data!? >How can I mirror a drive and have it work!!? (I assumed this was a >basic function of mirroring). >What is the correct downing and re-loading procedure for an SFTIII >server set. Under SFT III, the files you access are the files on your PRIMARY server, and not the mirrored servers files. Both disks and servers are seen as one unit via the MsEngine, even tho under Rconsole you see them as seperate entities, you are actually accessing the same "logical" server. All files read, are read from the primaries disk system, and all writes are mirrored as per usuall mirroring only between servers. Disk reads are not increased under SFT as under Duplexing or mirroring standard. From what you describe, I'd say your problems are definetaly connected to your secondary server, where either the disk/disks or the controller/cable are going wrong, and that data is getting corrupted between the servers, or the secondary server is causing data corruption in some manner (NIC/MSL) The list is endless really.. This is very possible, and I would do something about your secondary fast. You yourself state that your secondary server is constantly crashing, something which does not help an SFT environment in the first place. Have you considered replacing it with another server of the same type and build ? This type of problem is extremely difficult to diagnose without having the setup directly there. Please also, look for my NCF files that I will be posting here in a few days, the settings there may help you in some manner as well. ------------------------------ Date: Sun, 8 Mar 1998 16:10:28 +0200 From: Mike Glassman - Admin Subject: SFT III ncf files - LONG As I said I would, I'mm sending the primary servers ncf files for those of you who wish to look thru them. For general info, both SFT servers are set up as follows : IBM PCServer 330, 256Mb Ram, 21GB RAID-5 disks in an internall bay, IBM SCSI FW Raid controller, Onboard Adaptec FW SCSI Controller, 1 Thomas Conrad FO MSL card, 1 Madge TR NIC, onboard VGA controller. Both servers are connected to a bay 150Mb Switch and each port on the switch is set to 4.2k (or thereabouts), and connected via FO cable to the MSL NICs. The files are as follows : msauto.ncf - Main autoexec file msstart.ncf - Main startup file ioauto1.ncf - IO Engine 1's autoexec ioauto2.ncf - IO Engine 2's autoexec iostart1.ncf - IO Engine 1's startup iostart2.ncf - IO Engine 2's startup The actuall file names are without the numbers, and except for the msauto.ncf files, each pair sits in the c:\nwserver directory on the disk, with msstart.ncf as well. msauto.ncf sits in the sys:system directory on both servers (obviously). The msstart.ncf is common to both servers, whereas the numbered files, may be different between servers (IO Engine servers) as far as driver loads go. All parameters are tuned according to our systems settings, so some may not be appropriate for your settings, such as the min/max unclaimed memory parts. As to the others, you can play as you like. There are pointers to some settings in the Novell TID's, but personally I'v found that they don't always suit all systems. Anyways, Hope this helps some of you with your systems. Remember, you don't need all the params we have set here, some are defaulted. Also, remember that under SFT III, you run servman from the MsEngine to set combined info, and from each IoEngine to set the IoEngine's info. The same goes as far as set params. What works in one environment, may not in another. MSSTART.NCF ----------- ; The following two lines were added by the IntranetWare Support Pack installation LOAD C:\NWSERVER\PK411.NLM ; Added in order to overcome the RPL not find server error pmunload tskzerfx SET Minimum Packet Receive Buffers = 1000 LOAD SCSICD.CDM LOAD SCSIHD.CDM ; Loads support for long file names and FTP server support load LONG load NFS MSAUTO.NCF -------------------- set Time Zone = MET-2 set Daylight Savings Time Offset = 1:00:00 set Start Of Daylight Savings Time = (MARCH 21 00:00:01 AM) set End Of Daylight Savings Time = (SEPTEMBER 14 00:00:01 AM) set Default Time Server Type = SECONDARY # Note: The Time zone information mentioned above # should always precede the MS ENGINE name. set Bindery Context = O_IAA;.OU_KLALI.O_IAA; msengine name NS_IAA1 msengine ipx internal net 20100101 mount all load btrieve load bspxcom load conlog maximum=100 ; Network driver LOADs and BINDs are initiated via ; INITSYS.NCF. The actual LOAD and BIND commands ; are contained in INITSYS.NCF and NETINFO.CFG. ; These files are in SYS:ETC. sys:etc\initsys.ncf set dstrace=on load tsa410 load tsands load pserver .ps_ns_iaa1.o_iaa load monitor set tts backout file truncation wait time = 960 set maximum record locks per connection = 100000 set maximum packet receive buffers = 1500 SET Concurrent Remirror Requests = 32 load spxconfg i=10000 # Command to start LAN WorkGroup, NFS, and other products. ;UNISTART.NCF ;load unicon SET Remirror Block Size = 8 SET Server Failure Notification Name = .ADMIN.O_IAA SET Compression Daily Check Stop Hour = 4 SET Maximum Service Processes = 65 SET Minimum Service Processes = 30 IOAUTO1.NCF - Primary IO Engines ioauto.ncf file --------------------- load tcpip rip=no forward=yes bind ip to IP-TOKEN-1 addr=172.17.253.101 mask=FF.FF.0.0 load remote iaa1 load rspx load spxconfg i=10000 search add sys:adsm load dsmc sched -pass=some-password IOAUTO2.NCF - Secondary IO Engines ioauto.ncf file --------------------- load tcpip rip=no forward=yes bind ip to IP-TOKEN-1 addr=172.17.253.102 mask=FF.FF.0.0 load remote iaa1 load rspx load spxconfg i=10000 IOSTART1.NCF - Primary IO Engines iostart.ncf file ---------------------- ; The following lines were moved and/or added by the IntranetWare Support Pack installation set new start address for unclaimed memory block = 64000768 set new end address for unclaimed memory block = 250002304 LOAD C:\NWSERVER\PK411.NLM ; End of modifications made by IntranetWare Support Pack installation ; Added in order to overcome the RPL not find server error pmunload tskzerfx SET RESERVED BUFFERS BELOW 16 MEG=200 ioengine name NS_IAA1_IO1 ioengine ipx internal net 20100102 LOAD IPSRAID.HAM SLOT=5 LOAD MADGEODI SLOT=1 PORT=A000 MSB FRAME=TOKEN-RING NAME=IPX-TOKEN-1 LOAD MADGEODI SLOT=1 PORT=A000 MSB FRAME=TOKEN-RING_SNAP NAME=IP-TOKEN-1 BIND IPX IPX-TOKEN-1 NET=17217 LOAD TCMSL.MSL mem=d0000 int=5 set use diagnostic responder to validate lan functionality = on set ipx internet down wait time = 8 set MSL deadlock wait time = 10 set MSL error wait time = 3 set secondary take over wait time = 4.0 set check LAN extra wait time = 10 SET Minimum Packet Receive Buffers = 500 SET Maximum Packet Receive Buffers = 1500 IOSTART2.NCF - Secondary IO Engines iostart.ncf file --------------------- ; The following lines were moved and/or added by the IntranetWare Support Pack installation set new start address for unclaimed memory block = 64000768 set new end address for unclaimed memory block = 250002304 LOAD C:\NWSERVER\PK411.NLM ; End of modifications made by IntranetWare Support Pack installation ; Added in order to overcome the RPL not find server error pmunload tskzerfx SET RESERVED BUFFERS BELOW 16 MEG=200 ioengine name NS_IAA1_IO2 ioengine ipx internal net 20100103 LOAD IPSRAID.HAM SLOT=5 LOAD MADGEODI SLOT=1 PORT=A000 MSB FRAME=TOKEN-RING NAME=IPX-TOKEN-1 LOAD MADGEODI SLOT=1 PORT=A000 MSB FRAME=TOKEN-RING_SNAP NAME=IP-TOKEN-1 BIND IPX IPX-TOKEN-1 NET=17217 LOAD TCMSL.MSL mem=d0000 int=5 set use diagnostic responder to validate lan functionality = on set ipx internet down wait time = 8 set MSL deadlock wait time = 10 set MSL error wait time = 3 set secondary take over wait time = 4.0 set check LAN extra wait time = 10 SET Minimum Packet Receive Buffers = 500 SET Maximum Packet Receive Buffers = 1500 ------------------------------ Date: Thu, 12 Mar 1998 13:19:35 +0100 From: Jan Chochola Subject: Re: command line parameters to .NCF batch? >>Is it possible to pass command line parameters to a .NCF batch file? >>How do I reference the parameters from within the .NCF file. Yes, it is possible. Here is an example: File MYLOAD.NCF contains one line: load %1 At the [servername]: prompt, type: myload monitor ------------------------------ Date: Fri, 13 Mar 1998 09:38:42 -0800 From: Randy Richardson Subject: Re: DatbaseServer for NetWare >>I want a database server for run on a NetWare server with >>DOS, win 3.1 and win 95 clients. The following URLs will take you to an index of software manufacturers who have database products for NetWare: Home page http://netware.novell.com/database/ Database index http://NetWare.Novell.com/database/dbprof.htm ------------------------------ Date: Fri, 20 Mar 1998 16:33:01 +0100 From: Hans Nellissen Subject: Re: Unzip utility >Does anyone know of an unzip or similar utility which runs from the >console of a Netware 3.12 server ? Take a look at: http://www.jcsinc.com/ascentf.html There is an multizip.nlm (eval possible). ------------------------------ Date: Wed, 25 Mar 1998 15:13:54 -0800 From: Randy Richardson Subject: Re: console games >SNIPES.NLM Snipes is a multiuser (up to 5 players) text-based network game where you would run around in a maze the filled the entire screen, and it works with IntranetWare 4.11. It comes from the days of NetWare 2 for testing purposes - it was supposed to create a lot network traffic and work the system a little bit by constantly locking and unlocking records for every move. The filenames were actually NSNIPES.EXE (for monochrome monitors) and NCSNIPES.EXE (for color monitors). Later, NetWare Lite included a slightly improved version called "NLSNIPES.EXE" that correctly detected the display (the [Esc] key could be used to exit the game - previous versions required the use of "Ctrl-Break"). An enhanced stand-alone version of this game can be downloaded from one of the following URIs: http://bauxite.apricot.co.uk/ftp/bbs/area37/snipes.zip http://www.apricot.co.uk/ftp/bbs/area37/snipes.zip You'll have to ask Novell to send a copy of NLSNIPES.EXE to you, unless someone else knows where to find it? ------------------------------