Re: How to change /udev/rules.d/ for hsi
Thank you, that worked. I haven't read anything that mentioned adding the lines prior to the online statement. Thanks, Mark Vandale System Administrator Leader, MCS-MF z/VM, z/VSE & z/Linux Office : 860-823-2756 Cell : 860-705-1657 Textmsg : 8607051...@txt.att.net CSC This is a PRIVATE message. If you are not the intended recipient, please delete without copying and kindly advise us by e-mail of the mistake in delivery. NOTE: Regardless of content, this e-mail shall not operate to bind CSC to any order or other contract unless pursuant to explicit written agreement or government initiative expressly permitting the use of e-mail for such purpose. From: Ursula Braun To: LINUX-390@vm.marist.edu Date: 12/08/2011 12:00 PM Subject:Re: How to change /udev/rules.d/ for hsi Sent by:Linux on 390 Port Mark, try to move the "buffer_count" and "checksumming" line before the "online" line. The qeth driver does not allow to change the buffer_count value in online state. Regards, Ursula Braun, IBM Germany On Thu, 2011-12-08 at 10:13 -0500, Mark D Vandale wrote: > Here's some more detail if that helps... The hipersocket address e000. > I've added the last 2 lines, 1 to increase the buffer cnt and the other to > turn off checksumming. Neither parameter changes during a reboot. > > lsqeth -p still shows a cnt of 16 for the device > > -rw-r--r-- 1 root root 1296 2011-12-07 > 10:06 /etc/udev/rules.d/51-hsi-0.0.e000.rules > > # Configure hsi device at 0.0.e000/0.0.e001/0.0.e002 > ACTION=="add", SUBSYSTEM=="drivers", KERNEL=="qeth", IMPORT > {program}="collect 0.0.e000 %k 0.0.e000 0.0.e001 0.0.e002 qeth" > ACTION=="add", SUBSYSTEM=="ccw", KERNEL=="0.0.e000", IMPORT > {program}="collect 0.0.e000 %k 0.0.e000 0.0.e001 0.0.e002 qeth" > ACTION=="add", SUBSYSTEM=="ccw", KERNEL=="0.0.e001", IMPORT > {program}="collect 0.0.e000 %k 0.0.e000 0.0.e001 0.0.e002 qeth" > ACTION=="add", SUBSYSTEM=="ccw", KERNEL=="0.0.e002", IMPORT > {program}="collect 0.0.e000 %k 0.0.e000 0.0.e001 0.0.e002 qeth" > TEST=="[ccwgroup/0.0.e000]", GOTO="hsi-0.0.e000-end" > ACTION=="add", SUBSYSTEM=="ccw", ENV{COLLECT_0.0.e000}=="0", ATTR > {[drivers/ccwgroup:qeth]group}="0.0.e000,0.0.e001,0.0.e002" > ACTION=="add", SUBSYSTEM=="drivers", KERNEL=="qeth", ENV > {COLLECT_0.0.e000}=="0", ATTR > {[drivers/ccwgroup:qeth]group}="0.0.e000,0.0.e001,0.0.e002" > LABEL="hsi-0.0.e000-end" > ACTION=="add", SUBSYSTEM=="ccwgroup", KERNEL=="0.0.e000", ATTR {portno}="0" > ACTION=="add", SUBSYSTEM=="ccwgroup", KERNEL=="0.0.e000", ATTR {layer2}="0" > ACTION=="add", SUBSYSTEM=="ccwgroup", KERNEL=="0.0.e000", ATTR {online}="1" > ACTION=="add", SUBSYSTEM=="ccwgroup", KERNEL=="0.0.e000", ATTR > {buffer_count}="128" > ACTION=="add", SUBSYSTEM=="ccwgroup", KERNEL=="0.0.e000", ATTR > {checksumming}="no_checksumming" > > > Is there something else that needs to be changed ? SLES 11 > > > Thanks, > > Mark > > -- > For LINUX-390 subscribe / signoff / archive access instructions, > send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit > http://www.marist.edu/htbin/wlvindex?LINUX-390 > -- > For more information on Linux on System z, visit > http://wiki.linuxvm.org/ > -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For more information on Linux on System z, visit http://wiki.linuxvm.org/ -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For more information on Linux on System z, visit http://wiki.linuxvm.org/
How to change /udev/rules.d/ for hsi
Here's some more detail if that helps... The hipersocket address e000. I've added the last 2 lines, 1 to increase the buffer cnt and the other to turn off checksumming. Neither parameter changes during a reboot. lsqeth -p still shows a cnt of 16 for the device -rw-r--r-- 1 root root 1296 2011-12-07 10:06 /etc/udev/rules.d/51-hsi-0.0.e000.rules # Configure hsi device at 0.0.e000/0.0.e001/0.0.e002 ACTION=="add", SUBSYSTEM=="drivers", KERNEL=="qeth", IMPORT {program}="collect 0.0.e000 %k 0.0.e000 0.0.e001 0.0.e002 qeth" ACTION=="add", SUBSYSTEM=="ccw", KERNEL=="0.0.e000", IMPORT {program}="collect 0.0.e000 %k 0.0.e000 0.0.e001 0.0.e002 qeth" ACTION=="add", SUBSYSTEM=="ccw", KERNEL=="0.0.e001", IMPORT {program}="collect 0.0.e000 %k 0.0.e000 0.0.e001 0.0.e002 qeth" ACTION=="add", SUBSYSTEM=="ccw", KERNEL=="0.0.e002", IMPORT {program}="collect 0.0.e000 %k 0.0.e000 0.0.e001 0.0.e002 qeth" TEST=="[ccwgroup/0.0.e000]", GOTO="hsi-0.0.e000-end" ACTION=="add", SUBSYSTEM=="ccw", ENV{COLLECT_0.0.e000}=="0", ATTR {[drivers/ccwgroup:qeth]group}="0.0.e000,0.0.e001,0.0.e002" ACTION=="add", SUBSYSTEM=="drivers", KERNEL=="qeth", ENV {COLLECT_0.0.e000}=="0", ATTR {[drivers/ccwgroup:qeth]group}="0.0.e000,0.0.e001,0.0.e002" LABEL="hsi-0.0.e000-end" ACTION=="add", SUBSYSTEM=="ccwgroup", KERNEL=="0.0.e000", ATTR{portno}="0" ACTION=="add", SUBSYSTEM=="ccwgroup", KERNEL=="0.0.e000", ATTR{layer2}="0" ACTION=="add", SUBSYSTEM=="ccwgroup", KERNEL=="0.0.e000", ATTR{online}="1" ACTION=="add", SUBSYSTEM=="ccwgroup", KERNEL=="0.0.e000", ATTR {buffer_count}="128" ACTION=="add", SUBSYSTEM=="ccwgroup", KERNEL=="0.0.e000", ATTR {checksumming}="no_checksumming" Is there something else that needs to be changed ? SLES 11 Thanks, Mark -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For more information on Linux on System z, visit http://wiki.linuxvm.org/
How to change /udev/rules.d/ for hsi
I'm trying to update buffer size and turn off checksumming for hsi devices but the changes are taking affect. Is there something else that needs to be changed ? SLES 11 Thanks, Mark -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For more information on Linux on System z, visit http://wiki.linuxvm.org/
getting "device busy" on smbfs mounted area
In our fstab we mount a smbfs filesystem on z/OS.. It works for a while but at some point we are getting hung up. I have no idea what's causing this. Executing any cmd against it hangs for a while and the returns this device busy message. Is there a way to get around this or force it to be unmounted ? zlnx1a04:~ # umount /p2dev umount: /p2dev: device is busy Thanks, Mark Vandale -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: smbfs mount ? (or cifs)
I just received some additional information that I hadn't realized. Our Sles10 server is mounting this area from a SMB server on zOS, and it's that SMB that doesn't support the type "cifs", that's why I can't use it. Using the mount in fstab with type of smbfs, it works at boot up time. At some point later and I'm not sure how much later, the mount point gets hung up and stops responding, "Device busy". If I try to umount it I also get Device busy. What could be causing this ? Is it better to issue a "smbmount", is there any difference between mounting in fstab with type smbfs or issuing a straight smbmount cmd ? Thanks, Mark Vandale System Administrator Leader, MCS-MF z/VM, z/VSE & z/Linux Office : 860-823-2756 Cell : 860-705-1657 Textmsg : 8607051...@txt.att.net CSC This is a PRIVATE message. If you are not the intended recipient, please delete without copying and kindly advise us by e-mail of the mistake in delivery. NOTE: Regardless of content, this e-mail shall not operate to bind CSC to any order or other contract unless pursuant to explicit written agreement or government initiative expressly permitting the use of e-mail for such purpose. Mark Post To Sent by: Linux on LINUX-390@VM.MARIST.EDU 390 Port cc Subject Re: smbfs mount ? (or cifs) 03/19/2009 01:37 PM Please respond to Linux on 390 Port >>> On 3/19/2009 at 8:51 AM, Mark D Vandale wrote: > Is cifs supported on Suse sles10 x390 ? I had read that on Novell but > then someone had said to me it was not supported on MF Linux right now. I > did try it and received this in the messages log where as the smbfs works > now. > > kernel: CIFS VFS: cifs_mount failed w/return code = -2 It works for me. # df -hT /mnt Filesystem TypeSize Used Avail Use% Mounted on //10.10.220.13/root cifs388M 180M 188M 49% /mnt Mark Post -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: smbfs mount ? (or cifs)
Is cifs supported on Suse sles10 x390 ? I had read that on Novell but then someone had said to me it was not supported on MF Linux right now. I did try it and received this in the messages log where as the smbfs works now. kernel: CIFS VFS: cifs_mount failed w/return code = -2 Thanks, Mark Vandale System Administrator Leader, MCS-MF z/VM, z/VSE & z/Linux Office : 860-823-2756 Cell : 860-705-1657 Textmsg : 8607051...@txt.att.net CSC This is a PRIVATE message. If you are not the intended recipient, please delete without copying and kindly advise us by e-mail of the mistake in delivery. NOTE: Regardless of content, this e-mail shall not operate to bind CSC to any order or other contract unless pursuant to explicit written agreement or government initiative expressly permitting the use of e-mail for such purpose. John Summerfield To Sent by: Linux on LINUX-390@VM.MARIST.EDU 390 Port cc Subject Re: smbfs mount ? 03/18/2009 07:26 PM Please respond to Linux on 390 Port Amelia Nilsson wrote: > It is possible to mount smbfs from fstab. From your line i see that you > have a ":" to much and some of the "/" changed to "\". The line should > look like this: > > \\ipaddres/dirname /dirname smbfs username=testid,password=testpw 0 0 > > And ofcourse, if you want it to automount at boot you change the last 0 > to 1. Here are two correct mount commands. Note that cifs is the successor to smbfs: mount.cifs coco\\profile /mnt/coco/ --verbose -o user=administrator,password=**,ip=10.0.0.253,domain=class.corridors mount -t cifs coco\\software /mnt/coco.software --verbose -o user=administrator,password=**,ip=10.0.0.253,domain=class.corridors This works too: mount.cifs //coco/profile /mnt/coco/ -o user=administrator,password=**,ip=10.0.0.253,domain=class.corridors The arguments on the mount commands here should appear in fstab if that's what you are using. The credentials can be picked up from a file, so as to maintain confidentiality. Mostly, I prefer to not use backslashes so as to avoid hassles with programs that think they should be escaped and so reduce two to one, potentially several times. Note, on SLES10 (and probably 9) you should be specifying cifs, not smbfs; smbfs hasn't been maintained for some time. > > Best regards, > > - > Amelia Nilsson > > www.linuxchick.se > ame...@linuxchick.se > +46 762 77 42 77 > > > If you're not the intended recipient of this e-amail please delete this > e-mail and inform the sender. During previously stated circumstances > you're not allowed to copy or forward this e-mail. > > > On Wed, 2009-03-18 at 14:18 -0400, Mark D Vandale wrote: >> Can smbfs be mounted from /etc/fstab ? I can issue the cmd "mount -f >> smbfs //ipaddress/dirname /dirname username=testid,password=testpw" and >> get the smbfs mounted but I can't get this mounted by adding a line in >> the /etc/fstab . The format in the /etc/fstab is more like --> >> //ipadress:/dirname /dirname smbfs ro,username=testid,password=testpw 0 >> 0 Is this possible from /etc/fstab or do I need to do this from somewhere >> else ? This is SUSE SLES10. >> >> Thanks, >> >> Mark >> >> >> This is a PRIVATE message. If you are not the intended recipient, please >> delete without copying and kindly advise us by e-mail of the mistake in >> delivery. >> NOTE: Regardless of content, this e-mail shall not operate to bind CSC to >> any order or other contract unless pursuant to explicit written agreement >> or government initiative expressly permitting the use of e-mail for such >> purpose. >> >> -- >> For LINUX-390 subscribe / signoff / archive access instructions, >> send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit >> http://www.marist.edu/htbin/wlvindex?LINUX-390 >> > > -- > For LINUX-390 subscribe / signoff / archive access instructions, > send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit > http://www.marist.edu/htbin/wlvindex?LINUX-390 -- Cheers John -- spambait 1...@coco.merseine.nu z1...@coco.merseine.nu -- Advice http://webfoot.com/advice/email.top.php http://www.catb.org/~esr/faqs/smart-que
Re: smbfs mount ?
So the 2 preceding // before ipaddress is required for smbfs, we missed that I have 2 other lines prior to this one for nfs areas with no preceding // and they work. I did add these // and rebooted, it did work. Thanks ! Thanks, Mark Vandale System Administrator Leader, MCS-MF z/VM, z/VSE & z/Linux Office : 860-823-2756 Cell : 860-705-1657 Textmsg : 8607051...@txt.att.net CSC This is a PRIVATE message. If you are not the intended recipient, please delete without copying and kindly advise us by e-mail of the mistake in delivery. NOTE: Regardless of content, this e-mail shall not operate to bind CSC to any order or other contract unless pursuant to explicit written agreement or government initiative expressly permitting the use of e-mail for such purpose. Mark Post To Sent by: Linux on LINUX-390@VM.MARIST.EDU 390 Port cc Subject Re: smbfs mount ? 03/18/2009 04:23 PM Please respond to Linux on 390 Port >>> On 3/18/2009 at 4:12 PM, Mark D Vandale wrote: -snip- > This format looks like what > everyone has specified except the 0 0 at the end. No, it doesn't. //ipaddress/p2dev /pdev smbfs credentials=credfile,uid=emxpkid,gid=emeddev,fmask=774,dmask=774,rw 0 0 Would be the correct format. > I had found this > message in one of the logs and wasn't sure if it was related to this > statement in /etc/fstab: > In var/log/messages that says "kernel: smbfs: module not supported by > Novell, setting U taint flag. No, that just means the smbfs kernel module has been loaded, and it's one that (for whatever reason) isn't going to be supported by Novell if you have a problem with it, or if Novell support thinks that it is causing another problem elsewhere. Mark Post -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: smbfs mount ?
This is the format in the last /etc/fstab I tried. I started out using the credentials option and changed to specify the username and password just to make sure I didn't have something wrong with the file that contained the username and pw. The uid, gid, fmask and dmask was added this last time as well, before I had not specified those. This format looks like what everyone has specified except the 0 0 at the end. I had found this message in one of the logs and wasn't sure if it was related to this statement in /etc/fstab: In var/log/messages that says "kernel: smbfs: module not supported by Novell, setting U taint flag. ipaddress/p2dev /p2dev smbfs username=xxx,password=xxx,uid=emxpkid,gid=emeddev,fmask=774,dmask=777,rw 0 0 Thanks, Mark Vandale System Administrator Leader, MCS-MF z/VM, z/VSE & z/Linux Office : 860-823-2756 Cell : 860-705-1657 Textmsg : 8607051...@txt.att.net CSC This is a PRIVATE message. If you are not the intended recipient, please delete without copying and kindly advise us by e-mail of the mistake in delivery. NOTE: Regardless of content, this e-mail shall not operate to bind CSC to any order or other contract unless pursuant to explicit written agreement or government initiative expressly permitting the use of e-mail for such purpose. Mark Post To Sent by: Linux on LINUX-390@VM.MARIST.EDU 390 Port cc Subject Re: smbfs mount ? 03/18/2009 03:29 PM Please respond to Linux on 390 Port >>> On 3/18/2009 at 3:10 PM, Amelia Nilsson wrote: Actually it will work either way: //ipaddress/sharename or \\ipaddress/sharename Mark Post -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
smbfs mount ?
Can smbfs be mounted from /etc/fstab ? I can issue the cmd "mount -f smbfs //ipaddress/dirname /dirname username=testid,password=testpw" and get the smbfs mounted but I can't get this mounted by adding a line in the /etc/fstab . The format in the /etc/fstab is more like --> //ipadress:/dirname /dirname smbfs ro,username=testid,password=testpw 0 0 Is this possible from /etc/fstab or do I need to do this from somewhere else ? This is SUSE SLES10. Thanks, Mark This is a PRIVATE message. If you are not the intended recipient, please delete without copying and kindly advise us by e-mail of the mistake in delivery. NOTE: Regardless of content, this e-mail shall not operate to bind CSC to any order or other contract unless pursuant to explicit written agreement or government initiative expressly permitting the use of e-mail for such purpose. -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Re: EXAMINE 24: No Record Found detected - fatal error
The directory statements haven't changed since the initial install. The LVM which included 205 & 206 was configured during that install and was fine up until recently. In fact I have another copy of this system with the same configuration but not with these errors. Thanks, Mark Vandale Office: (860) 823-2756 Cell:(860) 705-1657 CSC This is a PRIVATE message. If you are not the intended recipient, please delete without copying and kindly advise us by e-mail of the mistake in delivery. NOTE: Regardless of content, this e-mail shall not operate to bind CSC to any order or other contract unless pursuant to explicit written agreement or government initiative expressly permitting the use of e-mail for such purpose. Rob van der Heij <[EMAIL PROTECTED] software.com> To Sent by: Linux on LINUX-390@VM.MARIST.EDU 390 Port cc <[EMAIL PROTECTED] IST.EDU> Subject Re: EXAMINE 24: No Record Found detected - fatal error 05/10/2006 02:31 AM Please respond to Linux on 390 Port <[EMAIL PROTECTED] IST.EDU> On 5/10/06, Mark D Vandale <[EMAIL PROTECTED]> wrote: > Any suggestions on how to fix these errors ? According to Linux, there's no data on that disk. It's not even formatted. You might want to do a DDR PRINT of the first tracks and see what's there. Since the device is recognized by the kernel but fails to find actual data, it could be someone just did an ICKDSF format of the first cylinder and labelled it. Also check slightly beyond - might be you now have cyl0 where before you had the disk from cyl 1 (as the 3338 may indicate). Maybe this device was unused before the upgrade? The LVM commands may reveal whether this physical device was indeed part of a volume group. Rob -- Rob van der Heij Velocity Software, Inc http://velocitysoftware.com/ -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
EXAMINE 24: No Record Found detected - fatal error
I'm running Suse SLES9 with SP3 on z/VM 5.1. Since I applied SP3 I have issued reboot often so I am not sure when these errors began. The system does come up and I can log in. As far as I can tell I don't see anything damaged on the system. This 0.0.206 is the second pack for a LVM. I'm not sure what's causing this or what to do. I found a patch patch-10870 (Security update for Linux kernel) that refers to some dasd errors but nothing that sounds like what I'm seeing here. Any suggestions on how to fix these errors ? coldplug scanning ccw: dasd(eckd): 0.0.0206: 3390/0A(CU:3990/01) Cyl:3338 Head:15 Sec:224 dasd(eckd): 0.0.0206: (4kB blks): 2403360kB at 48kB/trk linux disk layout dasde:<3>dasd_erp(3990): 0.0.0206: EXAMINE 24: No Record Found detected - fatal error dasd(eckd): I/O status report for device 0.0.0206: dasd(eckd): in req: 0e5f3ee8 CS: 0x40 DS: 0x0E dasd(eckd): device 0.0.0206: Failing CCW: 0e5f3fd0 dasd(eckd): Sense(hex) 0- 7: 00 09 04 01 b3 00 00 00 dasd(eckd): Sense(hex) 8-15: 86 00 01 00 0c 00 00 00 dasd(eckd): Related CP in req: 0e5f3ee8 dasd(eckd): CCW 0e5f3fb8: 63400010 0E5F3FD0 DAT: 40C0 dasd(eckd): CCW 0e5f3fc0: 47400010 0E5F3FF0 DAT: 0681 dasd(eckd): CCW 0e5f3fc8: 86001000 0E67A000 DAT: dasd_erp: 0.0.0206: (EXAMINE) ERP chain report for req: 0e5f3ee8 dasd_erp: 0.0.0206: 0e5f3ee8: c5c3d2c4 0e5c7868 end_request: I/O error, dev dasde, sector 0 Buffer I/O error on device dasde, logical block 0 Thanks, Mark Vandale Office: (860) 823-2756 Cell:(860) 705-1657 CSC This is a PRIVATE message. If you are not the intended recipient, please delete without copying and kindly advise us by e-mail of the mistake in delivery. NOTE: Regardless of content, this e-mail shall not operate to bind CSC to any order or other contract unless pursuant to explicit written agreement or government initiative expressly permitting the use of e-mail for such purpose. -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
test msg
This is just a test. No need to reply. -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
Hipersocket Frame size
Does anyone have any suggestions or guidlines on selecting a frame size in the definition of hipersockets ? I did read in the zSeries Hipersockets that the default Frame size is 16KB, compared with a maximum of 9KB for a Gigabyte Ethernet. But Hipersockets can also be defined as 24, 40 or 64KB. The selection depends on the data characteristics transported over a Hipersockets, which is a trade-off between performance and storage allocation. I want the best performance so what information do I need about the data to determine what frame size to define, and how much storage will I use if I define as 64KB ? All I have right now is that a client wants to transfer large amounts of data between z/OS and Linux but I don't have any details about how large. Is this main storage taken out of the total available on the processor or does it require more storage for each lpar to accomidate the hipersocket definitions ? I appreciate any suggestions. Thanks, Mark Vandale -- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390