Re: Strange zfcp problem on Redhat V5R2
On 30 July 2010 22:47, Sterling James ssja...@dstsystems.com wrote: Would someone be so kind, and post an example of this for us that are regex-challenged? For regular expressions I often cheat by using these 2 (windows) products developed by Jan Goyvaerts: RegexMagic RegexBuddy See his blogsite with product links here: http://www.just-great-software.com/blog.html I also have his coauthored book Regular Expressions Cookbook, see here: http://www.amazon.com/dp/0596520689?tag=jgsbookselectioncamp=213381creative=390973linkCode=as4creativeASIN=0596520689adid=132Z02HSJ4X8M03VNVXE; 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/
Re: building wireshark on SLES10SP2?
On 19 April 2010 21:56, Agblad Tore tore.agb...@volvo.com wrote: You don't have to use the z cpu cycles for wireshark: issue tcpdump into file in the zlinux machine copy the file to a linux desktop with opensuse for example run wireshark with that infile I think you missed the point of the thread. The whole point was to do dynamic, not post, investigation of traffic. Also the thread focus was on running wireshark on a PC not the mainframe. 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
Re: building wireshark on SLES10SP2?
On 16 April 2010 22:12, Richard Higson richard.hig...@gt.owl.de wrote: [5] http://wiki.wireshark.org/CaptureSetup/Pipes For anyone digging through this thread later, Richard points out [5] which is IMHO a great place to start. For Windows it contains an example in Python (and C) which is simple enough to script. 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
Re: building wireshark on SLES10SP2?
On 15 April 2010 08:04, Niels Horn niels.h...@gmail.com wrote: You can even monitor the traffic on-line using a Linux desktop with ssh access to your Linux server. I have never done this on Linux-390, but have on x86-based systems. The method is described in this post: http://blog.nielshorn.net/2010/02/using-wireshark-with-remote-capturing/ It is based on Slackware, but should work on most other Linux distributions. Nice Neils! I wonder what would be required for a Windows desktop that many of us are forced to use in corporations? Wireshark and netcat are available on windows, H'... 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
Re: Unable to kill process
Hi Ron, thats a support question if ever I heard one. Contact SAP/IBM. (Which of course will lead to a question on maintenace levels of linux and sap.) If kill -9 doesn't work then the process is in a system routine that can be cancelled. Are you saying that shutdown didn't work either and that you were forced to crash the linux guest? Did you have a problem with disks and/or nfs? (sap process needing to complete i/o?) were you alreday in shutdown? soemtimes the order that things are shutdoiwn is very important (users/sap/nfs/filesystem/network). mark On 6 April 2010 06:41, Ron Foster at Baldor-IS rfos...@baldor.com wrote: Hello, I just had to #CP logoff on of our production SAP systems. It had a SAP process on it with a process state of D interruptible sleep. The sap cancel options would not work. kill would not work. kill -9 would not work on it. Any ideas on how to avoid having to log off next time. Ron -- 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: Location of network configuration
If the heardware does not appear, then try using the hwup command using the hwcfg file name as a parameter in /etc/sysconfirg/hardware but dropping thw hwcfg prefix. 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
Re: swappiness drop_caches ?
Rob van der Heij wrote: On Thu, Apr 2, 2009 at 8:15 AM, Shane Ginnane sginn...@isi.com.au wrote: This is an indication of preference only - in a stress situation your input will be ignored. Well it should be - I have trouble explaining what Marcy saw at the setting of 60. My opinion* is that the two memory management algorithms interfere. There's probably Java code implementing various type of data cache classes in the JVM heap, likely on LRU basis. Linux will select candidates for swap-out also on LRU basis. This typically results in worst case scenario where the page selected for swap-out is the next one to be re-used by the application. When the page is re-used, the swap splot remains assigned until another swap-out for that page happens. So when this popular part of application memory just does not fit in what Linux set aside for it, you will see it over time cycle through. Resulting in every page eventually having a slot assigned in swap space. So it fills up swap space. When she lowered swappiness enough, it shifted the base line and left enough room for the popular part of the application without swap out the least recently used pages all the time. So it did not allocate all these swap slots and did not fill up swap space. *opinion: My best guess of what happened without measurements to prove it JAVA JVMs like real storage :-) The Garbage Collection (GC) runs periodically and references ALL JVM allocated storage. If Linux has paged out any of the JVM storage it must be brought back in for every GC cycle. This can be annoying if the JVM is not really using all of its allocated storage! There are 2 parms for the JVM: -XmssizeSet initial Java heap size -XmxsizeSet maximum Java heap size If you set the first one low (Xms) then the JVM will release unused memory down to this value, which is useful if you need the page frames for other tasks. GC will only reference the allocated storage, which could drop to this minimum value. swappiness is a control on how many *free* page frames linux is going to try an keep - if necessary by periodically paging some pages out. So if you are going to have a higher value for swappiness and run java, consider the above. 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
Re: s390 repositories for RHEL
Erling Ringen Elvsrud wrote: Hello list, I need a few packages that is not included in RHEL. First and foremost Puppet and dependencies which is at least ruby-rpm and ruby-shadow if I remember correctly. I have not found any s390 packages in the well known repositories like EPEL, rpmforge or dag. Do you know any repositories that contain s390 binary packages for RHEL? I could build them myself from the EPEL srpms, but if binary packages (from a reasonable source) exists I prefer to use that. Interestingly (maybe not for you), SLES-11 has a puppet rpm: Name: puppet Relocations: (not relocatable) Version : 0.24.5Vendor: SUSE LINUX Products GmbH, Nuernberg, Germany Release : 5.6 Build Date: Wed 25 Feb 2009 18:11:34 CET Install Date: (not installed) Build Host: s390z18 Group : Productivity/Networking/System Source RPM: puppet-0.24.5-5.6.src.rpm Size: 1624845 License: GPL v2 or later Signature : RSA/8, Wed 25 Feb 2009 18:11:38 CET, Key ID e3a5c360307e3d54 Packager: http://bugs.opensuse.org URL : http://reductivelabs.com/projects/puppet/ Summary : A network tool for managing many disparate systems Description : Puppet lets you centrally manage every important aspect of your system using a cross-platform specification language that manages all the separate elements normally aggregated in different files, like users, cron jobs, and hosts, along with obviously discrete elements like packages, services, and files. Distribution: SUSE Linux Enterprise 11 Name: puppet-serverRelocations: (not relocatable) Version : 0.24.5Vendor: SUSE LINUX Products GmbH, Nuernberg, Germany Release : 5.6 Build Date: Wed 25 Feb 2009 18:11:34 CET Install Date: (not installed) Build Host: s390z18 Group : Productivity/Networking/System Source RPM: puppet-0.24.5-5.6.src.rpm Size: 22708License: GPL v2 or later Signature : RSA/8, Wed 25 Feb 2009 18:11:38 CET, Key ID e3a5c360307e3d54 Packager: http://bugs.opensuse.org URL : http://reductivelabs.com/projects/puppet/ Summary : A network tool for managing many disparate systems Description : Puppet lets you centrally manage every important aspect of your system using a cross-platform specification language that manages all the separate elements normally aggregated in different files, like users, cron jobs, and hosts, along with obviously discrete elements like packages, services, and files. Distribution: SUSE Linux Enterprise 11 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
Re: Access linux files from CMS
Bernie Wu wrote: We have 3 separate environments, test, dev/qa and prod. Each environment uses different device addresses for the network configuration, 0.0.a04 for test, 0.0.a64 for dev/qa and 0.0.a60 for prod. But why do the different environments _have_ to have different VDEVs, I can understand that they may well require different RDEVs. Each environment could use the same VDEV but point it to a different LAN/RDEV/VSWITCH etc. 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
Re: SLES10 Kernels used on SAP application servers
Ron Foster at Baldor-IS wrote: Hello listers, We are a SAP shop. We are at SLES10 SP2. On our SLES10 SP2 application servers we have been using the kernel that came with SLES10SP2. That is: 2.6.16.60-0.21-default The SAP OSS Note (81737) that lets you know what release levels are supported says that this Service Pack level and kernel is certified. But then there is the sentence in that OSS Note that says A higher-level Linux kernel on top of a certified SP might be used until another released Linux kernel or released SP is available. My questions are 1) Do people stay at the kernel mentioned in the OSS note or 2) do they use the above sentence as permission to go to a higher kernel level? SAP Certification is only considered necessary for major kernel (and glibc) changes, minor kernel releases released for update by Novell/Suse are OK to use, but of course go through your own test phase before jumping to production as you would for any product change. SAP Certification is done for new SP's and new Versions of SLES. 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
Re: cpuplugd s390-tools - adjusting linux guest size dynamically
Rob van der Heij wrote: I only looked at cpuplugd for dealing with virtual processors, and my comments about waste of energy applied to that. Your energy or the System z's ? cpuplugd and z/VM 5.4 work nicely together taking idle processors out of the dispatcher list, and working with the Guest's SHARE. 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
Re: cpuplugd s390-tools - adjusting linux guest size dynamically
Rob van der Heij wrote: On Tue, Feb 17, 2009 at 2:02 PM, Mark Perry rita.co@googlemail.com wrote: Rob van der Heij wrote: I only looked at cpuplugd for dealing with virtual processors, and my comments about waste of energy applied to that. Your energy or the System z's ? cpuplugd and z/VM 5.4 work nicely together taking idle processors out of the dispatcher list, and working with the Guest's SHARE. Why would idle virtual processors be in the dispatch list? Spin locks ;-) 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
Re: LPAR to LPAR communication
Bernie Wu wrote: Hi Listers, We have a 2 LPARS, each hosting one VM, which in turn hosts several Linux guests. I would like to be able to query the number of guests on another LPAR from a linux guest on a different LPAR. Is this possible and if so, what do I have to do to set it up ? If linuxa1 is on one VM, and linuxb1 is on the other VM, and both guests have ssh (using keys is nicer) and vmcp installed and VM commands are authorized, then... linuxa1 #: ssh linuxb1 vmcp q users or linuxa1 #: ssh linuxb1 vmcp q n You could even use pipes and grep, perl or whatever you need to display the results in a format you required. 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
Re: Minidisks and DASD model 3/9
Rich Smrcina wrote: It is certainly safe to use model 9's for Linux minidisks. What you may want to consider is to split the minidisk up using LVM into separate filesystems for /var, /tmp and /usr (or some variation). Theres recent info on this in redbooks and redpiece and discussion in the archive. When using LVM2 you have the option to stripe the Logical Volume (LV), which contains the filesystem, across a number of physical volumes (PV) or DASD in stipesize chunks. This could offset any I/O queuing problems you may be concerned about with larger models of DASD, depending on the type and frequency of data accessed. 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
Re: z/VM CTCA won't start
Robin Atwood wrote: I asked him if he could get me a job in the region! BTW, I solved the CTC problem. -- -- Robin Atwood, Bangkok, Thailand. -- Get to the back of the queue(line)! Mark Perry, Uttaradit, Thailand. -- 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: HiperSocket Performance
Rakoczy, Dave wrote: zLinux assigns the MTU size according to the IQD CHPID definition. For sake of discussion lets say I set the CHPID to a Max Frame Size of 64K, that would give me an MTU size of 56K according to the Doc. Where can I control the size of the packets I'll send across the interface? In the Tape Blocksize / Record length as alluded to in the Adam's previous note? Sorry for all the questions... But I've got to learn this stuff somewhere. Hi Dave, others had already, carefully, suggested increasing the Frame-Size/MTU which may buy you some performance depending on the application's TCP/IP usage (streaming/transactional, etc.) We have seen in testing that the 8KB MTU is optimal for a wide range of application types. But you may want to tune for the absolute most bang-for-the-buck, which usually only works if you dedicate to one application type. Once you get past setting your MTU, you should then consider TCP/IP tuning: On z/OS TCPCONFIG TCPRCVB xx TCPSENDB xx On z/Linux in /etc/sysctl.conf: net.ipv4.tcp_rmem = net.ipv4.tcp_wmem = (set via the sysctl command, read man page) Also in the config files for your interfaces you can increase the number of QDIO buffers. /etc/sysconfig/hardware/hwcfg-qeth-bus-ccw-0.0. add a line like: QETH_OPTIONS=buffer_count=xxx The above can be verified using lsqeth. I have not given any exact values because everyones configurations (applications, hardware) will vary, suffice to say that the default values for all of the above are on the low side if you are aiming for higher performance. Also take care because these values will effect *ALL* of your TCP/IP applications not just the new one you are implementing. 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
Re: Losing IP connection on Virtual switch durin SLES10 installation
Robert J Brenneman wrote: sarcasm What - you mean everyone doesn't have 8 Gig of memory on their servers? /sarcasm qu'ils mangent de la brioche. 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
Re: Linux under z/VM performance document.
Dave Jones wrote: One of our leading z/VM and Linux under z/VM performance experts, Barton Robinson, has written a very nice performance overview and set of guidelines/best practices document for getting the most bank for the buck when deploying Linux as a guest of z/VM. The article can be found here: http://zjournal.com/index.cfm?section=articleaid=279 I'd suggest saving a copy locally someplace, so it can bve referred to at a later time. IMHO its more of an historical text than a reference document to be saved and referred to later. Its objective seems to be: Buy a good z/VM performance monitor That is a very good suggestion, but I might add: Ensure that your z/Linux and z/VM are at a current level to avoid almost all of the problems mentioned in this document. Such as SLES-10-SP2 or RHEL 5.2, and z/VM 5.3/5.4. 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
Re: Linux under z/VM performance document.
Barton Robinson wrote: Mark, don't be silly. You obviously don't follow the performance threads here when people go by the book and experience VERY BAD performance with CURRENT LEVELS. Barton, Oh I do follow the threads, sometimes just lurking. Example: Regarding MEMORY CP SET SRM STORBUF xxx xxx xxx CMM1 CMMA Regarding DISK STORAGE 3390 I/O bottlenecks - use 3390-3 or small minidisks Usage of LVM2 striping Regarding CPU Number of Guest Virtual CPs should be less that number of CPs of z/VM LPAR Don't see that discussed/recommended in your article? For a reference article or cookbook (which was Dave Jones point in the initial thread email) I would expect more tips. Such tips are much more useful for newcomers, than a history lesson of all the pain we have all been through over the years. Your article is good and covers many of the problems that we have all seen from SLES7 through to SLES9 and older z/VM releases (e.g 2GB). There are some timeless tips such as the use of VDISK. Please don't think I was criticizing your article directly it is actuate but many (not all) of the points you make pertain to older releases, and as I said I agree 100% with the use of a good z/VM performance monitor. Merry Christmas and a Very Happy New Year to you and yours 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
Re: VIPA and OSPFD setup
Hello Offerbaruch, I now have quagga+ospfd up and running on RHEL 5.2 (+ yum update ran today), and chatting happily with my SLES-10-SP2 systems ;-) First problem was that RHEL 5.2 starts a firewall automatically on boot. This blocks incoming OSPF packets, duh! I simply disabled the firewall (chkconfig --del iptables). Next problem (at least for me because I want a mixed environment of SUSE and RHEL) was the level of quagga on RHEL of 98.6 built in April 2007! SUSE SLES-10-SP2 ships quagga at 99.9 (and even 99.7 on SLES-9-SP4). I took the source rpm from the source DVD for 98.6, downloaded the latest 0.99.11 quagga tarball from quagga.net, reworked the spec file did an rpmbuild -bb, then installed my nice new quagga rpm ;-) Rather than fill this mailing list with attachments - if you want anything contact me offlist: rita.co@googlemail.com regards mark perry Mark Perry wrote: Hi offenbach, I work on a day-to-day basis with quagga on SLES-10-SP2 and OMPROUTE on z/OS (1.9/1.10). I really would like to help you as I have just installed RHEL 5.2 and intend to install both RHEL 5.3 and SLES-11 over the coming months. I will need to include all of these systems into our existing High-Availability vipa/ospf configuration. The catch is that I am traveling all of this week and won't be able to look into this properly until next Monday 22nd. In the mean time perhaps someone else will assist you, if not I'll be checking my email next week - if you want to do this offlist my email is: rita.co@googlemail.com regards Mark Perry -- 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: VIPA and OSPFD setup
Hi offenbach, I work on a day-to-day basis with quagga on SLES-10-SP2 and OMPROUTE on z/OS (1.9/1.10). I really would like to help you as I have just installed RHEL 5.2 and intend to install both RHEL 5.3 and SLES-11 over the coming months. I will need to include all of these systems into our existing High-Availability vipa/ospf configuration. The catch is that I am traveling all of this week and won't be able to look into this properly until next Monday 22nd. In the mean time perhaps someone else will assist you, if not I'll be checking my email next week - if you want to do this offlist my email is: rita.co@googlemail.com regards Mark Perry עופר ברוך wrote: Hi all, I am trying to setup a VIPA interface on my RHEL V5.2 and distribute the routes using ospfd (using the quagga package). What I am trying to accomplish is having 3 physical interfaces (ctc0, hsi0 and eth0) and one dummy interface representing the VIPA interface. The main reason for doing this is getting z/OS machines to use hsi device when connecting to the VIPA address and the rest of the world should connect using the eth interface. Furthermore in the case of an hsi failure I would like the z/OS to automatically use the eth interface in order to connect to the Linux machine. I really don't want to use static routes so I went to OSPF to find the answer. Can anyone please send a zebra.conf and ospfd.conf example to accomplish that? This are the files I created (obviously there is something wrong there� I know very little about ospf but I did my best J) zebra.conf file: hostname LNXS3 password zebra # log file /var/log/quagga/zebra.log information interface eth0 no shutdown ospfd.conf file: ! -*- ospf -*- ! ! OSPFd sample configuration file ! ! hostname ospfd password zebra log file /var/log/quagga/ospfd.log !enable password please-set-at-here ! interface eth0 ip ospf priority 1 !interface hsi0 !ip ospf priority 2 interface ctc0 ip ospf priority 2 ! interface dummy0 router ospf redistribute connected passive-interface dummy0 network 172.20.140.0/24 area 0.0.0.0 network 172.20.101.1/32 area 0.0.0.0 network 10.10.3.0/24 area 0.0.0.0 network 10.10.2.0/24 area 0.0.0.0 this is my ifconfig output: ctc0 Link encap:Serial Line IP inet addr:10.10.3.1 P-t-P:10.10.3.2 Mask:255.255.255.255 UP POINTOPOINT RUNNING NOARP MTU:32760 Metric:1 RX packets:210 errors:0 dropped:0 overruns:0 frame:0 TX packets:224 errors:16 dropped:16 overruns:0 carrier:16 collisions:0 txqueuelen:100 RX bytes:15008 (14.6 KiB) TX bytes:16532 (16.1 KiB) dummy0Link encap:Ethernet HWaddr 92:C9:FC:44:84:11 inet addr:172.20.101.1 Bcast:172.20.101.3 Mask:255.255.255.252 inet6 addr: fe80::90c9:fcff:fe44:8411/64 Scope:Link UP BROADCAST RUNNING NOARP MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 b) TX bytes:0 (0.0 b) eth0 Link encap:Ethernet HWaddr 02:00:00:00:00:03 inet addr:172.20.140.13 Bcast:172.20.140.255 Mask:255.255.255.0 inet6 addr: fe80::200:0:200:3/64 Scope:Link UP BROADCAST RUNNING NOARP MULTICAST MTU:1492 Metric:1 RX packets:2923 errors:0 dropped:0 overruns:0 frame:0 TX packets:1562 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:242871 (237.1 KiB) TX bytes:201379 (196.6 KiB) hsi0 Link encap:Ethernet HWaddr 00:00:00:00:00:00 inet addr:10.10.2.36 Bcast:10.10.2.255 Mask:255.255.255.0 inet6 addr: fe80::200:ff:fe00:0/64 Scope:Link UP BROADCAST RUNNING NOARP MULTICAST MTU:16384 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:195 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:0 (0.0 b) TX bytes:12256 (11.9 KiB) loLink encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:648 errors:0 dropped:0 overruns:0 frame:0 TX packets:648 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:51461 (50.2 KiB) TX bytes:51461 (50.2 KiB) -- 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
Re: question on top
Barton Robinson wrote: Yes, top lies. So are you really saying a single threaded process is using more than one cpu? Don't need much more proof than that. Barton I don't think TOP shows threads by default, so a multithreaded process would so up as a single line. In TOP type H to show threads. Also you can type 1 to list all CPUs rather than a summary. mark -- 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
Re: Recover a root disk
On my x86 RedHat system I have two useful packages: testtool - which supplies photorec - homepage http://www.cgsecurity.org/wiki/TestDisk ext3undel - supplies scripts to recover deleted files via photorec - homepage http://www.izzysoft.de/ Can't see these for s390x, but perhaps now there is a market. mark -- 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
Re: different take on Java 6 question
Mark Post wrote: On 11/21/2008 at 10:07 AM, [EMAIL PROTECTED] wrote: found the IBM link for jdk downloads - now a different question. Will the 31-bit jdk work - and work well - in a 64-bit linux environment. We run the 32-bit java on 64-bit AIX without any difficulty since quite a few of our apps refuse to work with the 64-bit java. Don't have any experience in the linux world with this, though. The recommendation I've heard from IBM is to run 31-bit if at all possible. It's largely going to depend on the heap size you need for your application. Applications drive the java requirement - SAP requires the 64bit IBM SDK. This opens another ongoing peeve of mine... Novell-Suse ship the IBM javas on the DVD and supply updates - great job! Also nice work on the /etc/alternates and the update-alternatives command - all very nice if you need to run multiple levels of java at the same time. But these rpms install into /usr, the rpms from the IBM website install into /opt. SAP recommends that one should use the java from the IBM website because IBM often posts updates before Novell-Suse can which kind of defeats all the good work done by Novell-Suse :-( It would be nice if the two companies could sync-up on how the rpms should be packaged. mark -- 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
Re: VIPA + OSA + VLAN?
Mauro Souza wrote:
As now, I am starting VIPA on rc.local file, but I don't know if it's the
best place to start it. Any other options?
Hi Mauro,
I don't have a RHEL 4.6 system to verify, I have done this many times on
SLES 10.
in order to load the dummy module in the SLES /etc/sysconfig/kernel
script I have modified the following line:
MODULES_LOADED_ON_BOOT=vmcp dummy
On SLES 10, I simply define in /etc/sysconfig/network a file named
ifcfg-dummy0 as follows:
BOOTPROTO=none
UNIQUE=
STARTMODE=onboot
MTU=8992
IPADDR=10.101.4.216
NETMASK=255.255.255.255
NETWORK=10.101.4.216
BROADCAST=0.0.0.0
The dummy device is then brought up during runlevel 3.
To handle the qethconf in each OSA interface definition file
ifcfg-xxx I add the line:
POST_UP_SCRIPT=setvipa
Then in /etc/sysconfig/network/scripts I have added a script named
setvipa as follows:
#!/bin/bash
#
# This script is called via ifup when it
# processes an ifcfg--* script that contains:
# POST_UP_SCRIPT=setvipa
#
# First obtain interface name
INT=$2
#
# Now read in the dummy0 VIPA details
. /etc/sysconfig/network/ifcfg-dummy0
#
# Copy the IP address for dummy0
VIPA=${IPADDR}
#
# Set the VIPA address into the OSA interface
/sbin/qethconf vipa add ${VIPA} ${INT}
Maybe you, or someone from Redhat, can convert the above
mark
--
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
Re: VIPA + OSA + VLAN?
Mauro Souza wrote: Hi people! I am facing some problems setting up a network infraestructure. I had never worked with vlan + vipa before, so... Environment: zVM 5.3 and RHEL4.6. I have to set up two linuxes, and they have direct access to an OSA. We have 2 OSA interfaces, so I defined one interface for each, for the high availability sake. I am not using vswitch, only the direct attached OSA. First, I have defined the LAN: DEFINE LAN INTERNET OWNER SYSTEM TYPE QDIO IP UNRESTRICTED MAXCONN INFINITE Then, in USER DIRECT NICDEF 6000 TYPE QDIO LAN SYSTEM INTERNET NICDEF 7000 TYPE QDIO LAN SYSTEM INTERNET After booting both Linuxes, I configured the VLAN on both: vconfig eth0 add 100 vconfig eth1 add 100 linux1: ifconfig eth0.100 192.168.1.1 ifconfig eth1.100 192.168.1.2 linux2: ifconfig eth0.100 192.168.1.4 ifconfig eth1.100 192.168.1.5 So far, so good. Until here, I am able to ping both interfaces from both Linuxes. So I added VIPA: modprobe dummy linux1: ifconfig dummy0 192.168.1.3 qethconf vipa add 192.168.1.3 eth0 qethconf vipa add 192.168.1.3 eth1 linux2: ifconfig dummy0 192.168.1.6 qethconf vipa add 192.168.1.6 eth0 qethconf vipa add 192.168.1.6 eth1 As soon as I activate the dummy0 interface, a blackout strikes the network. Both linuxes can ping only their own interfaces, and cannot ping each other interfaces. I tried to create a route to VIPA thru the VLAN (e.g. route add -host 192.168.1.3 -dev eth0.100), but no luck. As soon as I destroy the VIPA (qethconf vipa del 192.168.1.6 eth0 ; qethconf vipa del 192.168.1.6 eth1 ; ifdown dummy0 ; rmmod dummy) I can ping again. I am not pretty sure if adding VIPA to eth0 instead of eth0.100 is the right thing to do. I tried to add a vipa to eth0.100, but qethconf complained about an illegal, incorrect and inexistent eth0.100. I am now thinking about changing the interface naming convention to use vlan100 instead of eth0.100... Someone have an advice? Hi Mario, rather than just giving up on this and switch to a vswitch solution as suggested by others (what happens if you want to run the guest in a native LPAR later for example?) Check the routing tables before and after adding the VIPA(s): ip route show mark -- 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
Re: VIPA + OSA + VLAN?
Mark Perry wrote: Mauro Souza wrote: Hi people! I am facing some problems setting up a network infraestructure. I had never worked with vlan + vipa before, so... Environment: zVM 5.3 and RHEL4.6. I have to set up two linuxes, and they have direct access to an OSA. We have 2 OSA interfaces, so I defined one interface for each, for the high availability sake. I am not using vswitch, only the direct attached OSA. First, I have defined the LAN: DEFINE LAN INTERNET OWNER SYSTEM TYPE QDIO IP UNRESTRICTED MAXCONN INFINITE Then, in USER DIRECT NICDEF 6000 TYPE QDIO LAN SYSTEM INTERNET NICDEF 7000 TYPE QDIO LAN SYSTEM INTERNET After booting both Linuxes, I configured the VLAN on both: vconfig eth0 add 100 vconfig eth1 add 100 linux1: ifconfig eth0.100 192.168.1.1 ifconfig eth1.100 192.168.1.2 linux2: ifconfig eth0.100 192.168.1.4 ifconfig eth1.100 192.168.1.5 So far, so good. Until here, I am able to ping both interfaces from both Linuxes. So I added VIPA: modprobe dummy linux1: ifconfig dummy0 192.168.1.3 qethconf vipa add 192.168.1.3 eth0 qethconf vipa add 192.168.1.3 eth1 linux2: ifconfig dummy0 192.168.1.6 qethconf vipa add 192.168.1.6 eth0 qethconf vipa add 192.168.1.6 eth1 As soon as I activate the dummy0 interface, a blackout strikes the network. Both linuxes can ping only their own interfaces, and cannot ping each other interfaces. I tried to create a route to VIPA thru the VLAN (e.g. route add -host 192.168.1.3 -dev eth0.100), but no luck. As soon as I destroy the VIPA (qethconf vipa del 192.168.1.6 eth0 ; qethconf vipa del 192.168.1.6 eth1 ; ifdown dummy0 ; rmmod dummy) I can ping again. I am not pretty sure if adding VIPA to eth0 instead of eth0.100 is the right thing to do. I tried to add a vipa to eth0.100, but qethconf complained about an illegal, incorrect and inexistent eth0.100. I am now thinking about changing the interface naming convention to use vlan100 instead of eth0.100... Someone have an advice? Hi Mario, rather than just giving up on this and switch to a vswitch solution as suggested by others (what happens if you want to run the guest in a native LPAR later for example?) Check the routing tables before and after adding the VIPA(s): ip route show mark Also 1) You are not directly using an OSA, you are accessing a (closed) z/VM GuestLAN 2) Syntax: vconfig add eth0 100 mark -- 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
Re: Lx86
Rich Smrcina wrote: Watch this space... In Q1 2009 Mantissa will deliver a system that permits unaltered Windows operating systems to run under z/VM. Using a desktop appliance running RDC, users will be able to connect to their virtual Windows images running in the VM environment. Goodbye desktop hardware, remote maintenance, high power consumption, machine order lead time. I just got to see the Microsoft Licensing for this one ;-) mark -- 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
Re: hipersocket
Ceruti, Gerard G wrote: Hi All When defining the QDIO CHPID in HCD is there any issue with defining a Frame Size of 64K ?, if we leave our MTU at 8K other than the some storage wastage are there any other issues we need to be aware of Regards Gerard Ceruti may the 'z' be with you Hi Gerard, it works without other issues. But I'm curious as to why you would want to set the MTU=8K? Jumbo frames are typically set on OSA's at 8992, so unless this is strictly LPAR to LPAR on the same CEC, then PATH MTU DISCOVERY may be in order (or setting all IFs to the same MTU) to prevent segmentation if packets arrive from a OSA and are bound for the HiperSockets. It all depends on what packet routing you will be doing. Without knowing your OS's, applications or network topology its hard to advise. mark -- 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
Re: hipersocket
Ceruti, Gerard G wrote: Hi Mark This is pure HiperSocket setup, all internal on the same CEC, RHEL and SUSE guests talking to DB2 on zOS. Regards Gerard Ceruti may the 'z' be with you Hi Gerard, I believe you are working with Jochen Roehig? If so he is in discussions with Volker Schoelles and I, and we can discuss in depth off list. mark -- 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
Re: NFS - z/OS and Linux
in addition the the good info that Stewart Thomas wrote, I would add that z/OS USS files can be tagged with a codepage. You can see this using ls -lT, and modify using the chtag command. We have USS filesystems mounted on zLinux (with default codepages in the mount command as Stewart documented) from a z/OS Server and the translation is automatic down to the individual file level. So even though we specify in the mount that translation should be from EBCDIC to ASCII I can also access files that are tagged as ASCII on the z/OS USS side. mark -- 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
Re: 127.0.0.2 in /etc/hosts?
Michael MacIsaac wrote: I still have never had my virtual network cable fall out of my virtual NIC or my virtual switch (sorry, couldn't resist :)) try from Linux: vmcp det nic xxx ;-) mark -- 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
Re: 127.0.0.2 in /etc/hosts?
Michael MacIsaac wrote: Alan, That smells like someone manually created it. 127.anything, not just 127.0.0.2, is localhost. No. I just reinstalled SLES 10 SP2 to test. On the Hostname and Domain Name panel in the second half of the install I uncheck Change Hostname via DHCP and leave Write Hostname to /etc/hosts checked. Same /etc/hosts on the new system: # tail -1 /etc/hosts 127.0.0.2 gpok149.endicott.ibm.com gpok149 This causes problems with applications like SAP, which make reverse and forward lookups to ensure it understands the network (security/sanity check maybe?). It expects the IP to be on one of the interfaces it is using for communication with other App Servers. For SAP the simple fix is just to comment the line out of /etc/hosts and fallback to DNS lookups. mark -- 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
Re: YAST fails to grow an ext3 FS on LVM.
Fargusson.Alan wrote: Has anyone ask Novell why YAST fails to grow a logical volume that has an ext3 filesystem on it? Alan, I see from this thread that you mention SLES-10-SP1. With this level you can only grow a 4KB bs filesystem to the next 16GB boundary. With SLES-10-SP2 any newly created filesystems are pre-configured to allow growth. Any filesystems created on SP1 will still have limitations even under SP2. I have not yet found any way to upgrade an SP1 filesystem without recreating it from scratch. The man page for ext2online mentions an ext2prepare command, but this is not shipped with SLES-10-SP2. Neither does tune2fs seem to cover this, its a mystery to me right now. It would be nice if SLES had a method to upgrade filesystems, even if this had to be done one-time offline. mark -- 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
Re: Root filesystem error switches to ReadOnly
Fargusson.Alan wrote: Unless this changed in 2.6 the block devices go through VFS, and the blocks get cached by the VFS layer. VFS only comes into play when a filesystem (on a block device) is mounted. If you access a block device directly (not via a path of the mountpoint) then VFS is not involved. The block device is a direct entry to the device driver, you are dealing with blocks and not filesystems at this level. The buffer cache may be used, but this is not filesystem aware, it just caches previously read/written blocks. The point remains that doing a dd of the /dev/dasdn while mounted can lead to a copy that is in a inconsistent state. mark -- 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
Re: Support dates for IBM products on new releases of SLES
I can't speak for the products Marcy has listed, but there are factors that often prevent support from day 1: 1) The schedule is Novell's and not IBM's 2) The Schedule can slip even for beta releases causing testing delays. 3) IBM Testing is not complete until you have tested on a Novell GA release. 4) There have been last minute changes from Novell that caused confusion about what is GA (SLES9 SP4 found a major bug which caused a delay.) While all major vendors will of course be involved in the Beta testing of any new Novell release, I would think it reasonable to assume that any vendor may not fully certify/support a new Novell GA release for at least a few weeks following the GA release date, 2-6 depending on problems found. mark Stephen Frazier wrote: Normally, for most products, IBM tests them on the new release before the availability date. If it runs with no problems then it is supported from day 1. If they have to change something then support may be later. If it is easy to fix they may support it (with the fix) from day 1. Marcy Cortes wrote: Have to do some 2009 planning... How long is it typically between the availability of the next release of SLES (or RHEL for that matter) before IBM will support WAS, DB2, MQ Series on it? Marcy -- 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
Re: Distribution ages, was: Linux version
Evans, Kevin R wrote: Well, for sure, some things stand out more when you look at the code on paper. Sometimes, scrolling up and down on a screen tends to obfuscate the code. This is more so (at least for me) with assembler code (which I'll admit to not knowing as well as C). Code? We had hex dumps about 8+ inches thick (today there would probably be a company policy about lifting so much weight)- just tear and twist a corner for a bookmark (maybe with comment), and use a highlighter pen for the interesting data on a page. Of course the backside of the continuous sheet output could be used for disassembly/comments/diagrams etc. ;-) Then we would work out an amaspzap fix. Kids today... mark -- 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
Re: List FAQ etc
Evans, Kevin R wrote: I don't believe that everyone believes in pruning out everything. Many people reply at the top. If for no other reason than replies can be kept in context. There are arguments for and against top or bottom replies. In this reply I have chosen below, but then I have pruned the whole email thread from your posting. There is the point, we are not exchanging emails, we are posting to a common thread within a mailing-list. When posting to a mailing list there is no need to re-post data that is already available within the mailing list. You only need to leave the relevant text that you wish to refer to. SNIPING is encouraged! With SNIPING and bottom replies you can read the whole post from top to bottom in a logical manner. If you can't recall the previous post, you can always refer back to the mailing list archives. mark -- 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
Re: List FAQ etc
Mary Anne Matyaz wrote: Actually, sniping is discouraged. Snipping is encouraged. :) http://wilk4.com/humor/doh.htm mark -- 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
Re: List FAQ etc
Evans, Kevin R wrote: Sometimes these posts can last for over a week...I can't always remember the original posted question. You do look at the archives by thread, right? example: http://www.mail-archive.com/linux-390@vm.marist.edu/ mark -- 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
Re: Root filesystem error switches to ReadOnly
Fargusson.Alan wrote: Doing a dd on a device goes through the cache, so in this case the cache can't be the problem. Really? dd is designed to work on block devices that are not mounted, as well as files within a mounted filesystem. I thought the VFS only cached mounted filesystems? If the dd is using something like if=/dev/dasda of=dev/dasdb then I would have thought that the cache would not be used. The cache is only used for mounted filesystems, i.e. files/blocks referenced within the FS via the mountpoint. mark -- 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
Re: Linux version
R P Herrold wrote: [EMAIL PROTECTED] ~]$ lsb_release -a LSB Version: \ :core-3.1-ia32:core-3.1-noarch:graphics-3.1-ia32:graphics-3.1-noarch Distributor ID: CentOS Description:CentOS release 5.2 (Final) Release:5.2 Codename: Final [EMAIL PROTECTED] ~]$ ihlscoh2:~ # lsb_release -a LSB Version: core-2.0-noarch:core-3.0-noarch:core-2.0-s390x:core-3.0-s390x:desktop-3.1-noarch:desktop-3.1-s390:desktop-3.1-s390x:graphics-2.0-noarch:graphics-2.0-s390:graphics-2.0-s390x:graphics-3.1-noarch:graphics-3.1-s390:graphics-3.1-s390x Distributor ID: SUSE LINUX Description:SUSE Linux Enterprise Server 10 (s390x) Release:10 Codename: n/a ihlscoh2:~ # What a wonderfully kept secret that has been. Many thanks for pointing this out Russ! mark -- 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
Re: Root filesystem error switches to ReadOnly
van Sleeuwen, Berry wrote: Hello list, For various tests we have created a group of new SLES10 guests. They all are just one 3390-3 volume so they can be cloned vary fast. If you use FLASHCOPY its fast regardless of DASD size. But for some reason these machines sometimes switch the rootdisk, and therefore the entire filesystem, to readonly. I have tried a remount but that didn't work so only a reboot will get the / back in write mode. The reboot itself run without errors. But even that sometimes doesn't solve the error and we have to clone the guest again. We also have a different SLES10 installation and there we don't see any errors there but then we aren't running any load there. I can't find any reason for this, other than what I can find in de guest console log. Any ideas? What can be the cause of this and how can it be fixed? You have a problem in with you root filesystem, so it switches to R/O to give you a chance to fix it, e.g. with fsck. The guest is a SLES10 SP2. I have many SP2 systems, all with ext3 - never any problems at all. [EMAIL PROTECTED]:~ uname -a Linux nlzlx921 2.6.16.60-0.21-default #1 SMP Tue May 6 12:41:02 UTC 2008 s390x s390x s390x GNU/Linux [EMAIL PROTECTED]:~ df Filesystem 1K-blocks Used Available Use% Mounted on /dev/dasda12365444 1682640562644 75% / udev14562064145556 1% /dev The message from the console: 08/08/19 12:23:31 NLZLX921 VMLXES21: EXT3-fs error (device dasda1): ext3_readdir: bad entry in directory #270514: rec_len is 08/08/19 12:23:31 NLZLX921 VMLXES21smaller than minimal - offset=0, inode=0, rec_len=0, name_len=0ŸAborting journal on device 08/08/19 12:23:31 NLZLX921 VMLXES21dasda1.Ÿext3_abort called.ŸEXT3-fs error (d 08/08/19 12:23:31 NLZLX921 VMLXES21: Aug 19 12:23:31 nlzlx921 syslog-ngÃ1092¨: io.c: do_write: write() failed (errno 30), Read-only 08/08/19 12:23:31 NLZLX921 VMLXES21file system When I asked one of our linux specialists he did mention there were a few bugs relating ext3 that have been fixed in RHEL 5.1. Now I have noticed that RHEL 5.1 is on a newer kernel level than the SLES 10 SP2. Could a bug in ext3 by any chance be the problem here? The bug is most likely in your cloning process, perhaps the DASD extents copied and/or used are different. This will only show up when data is used within the bad extent range. mark -- 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
Re: Root filesystem error switches to ReadOnly
van Sleeuwen, Berry wrote: Hello Mark, I meant fast in a way that the / is copied and then changes the hostname and IP address. No other fancy things, just a default environment for our support staff to play around with. The cloning process is loosly based on the cloning scripts from redp4322. The dasd is copied using DASDFMT and dd. Both source (installation machine) and targets are identical, other than the obvious different location of the minidisk extents. The / is a 3338 cylinder minidisk so there it even uses the same minidisk extents but on different DASD volumes. There is one thing I now see, I first regarded as an error due to new disks, but it is still present even when the disks were in use for the guest before. When the copy is done there are some IO errors. FLASHCOPY is still faster ;-) 08/08/19 13:41:20 NLZLX920 VMLXES21: end_request: I/O error, dev dasdd, sector 361776Ÿprintk: 630 messages suppressed.ŸBuffer I/O 08/08/19 13:41:20 NLZLX920 VMLXES21error on device dasdd, logical block 45222Ÿlost page write due to I/O error on dasddŸBuffer 08/08/19 13:41:20 NLZLX920 VMLXES21I/O error on device dasdd, logical bl The disk is linked, then DASDFMT and dd. It looks like the IO errors appear within the dd part of the copy process. But how to fix that? Should the guest have some mdisk option to avoid errors? Thanks, Berry. If doing a dd from linux and the minidisks are identical in size, then one must assume that you have the rights to write to all cylinders. If you issue a vmcp q v da - are source and target the same number of cylinders? I also assume that the source and target are not mounted to Linux while the dd is running? Have you done an fsck on the source volume to ensure it is good? Are you specifying any dd options? Can you do a foreground dd (with -v) and reproduce the error? Is z/VM logging anything - EREP? mark -- 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
Re: Distribution ages, was: Linux version
Evans, Kevin R wrote: Kinda makes one realize how long z/OS (or its ancestors) has been around, doesn't it? K And those of us who have worked on it too ;-) mark -- 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
Re: Root filesystem error switches to ReadOnly
van Sleeuwen, Berry wrote: Hello Mark, I'll have to look into the flashcopy process but have not taken the time for it just yet. I guess one thing that would be wrong then is that I copy from the live system. So the source nlzlx920 is being copied from within the nlzlx920. But would that introduce IO errors on the target disks? If anything I'd expect to see errors during boot of the target guest. Either in a filesystemcheck or dmesg afterwards. Much like when you DDR a live VM and get a warmstartdata error when the target is IPLled. Perhaps a rethink of the process is in order. I guess a resque system would be an option. I do like to have a resque machine so I could also use that machine to do the cloning. As for other questions, I also assume I have the rights. No errors on that part. Same amount of cylinders. I have not yet ran the fsck on the source nor did I test it in a failsafe IPL of a new target. I dd with dd bs=4096 if=$sDev of=$tDev no other options. Haven't ran dd during the process by hand, other than a manual test to clone the first machine and to test the steps in the script. No erep or errors within the operator log. There is no error from the VM part. It's only within the linux. Regards, Berry. Never clone a live Linux filesystem. Once the filesystem has been mounted the filesystem is cached in storage (as are updates) and the on disk filesystem is marked 'dirty. Stop the master linux system with a clean shutdown, and copy the disk using flashcopy or ddr if you prefer. This new copy is the master disk to be used for future copies. Reboot the Linux system and run an e2fsck on the new copy to ensure it is good. Then clone as you like ;-) mark -- 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
Re: Distribution ages, was: Linux version
Evans, Kevin R wrote: This is getting like Monty Python g. One guy says When we were young, we used to eat the leather from our shoes. The other guy says You had shoes?. OS/360 - some of you guys make me feel young, thanks ;-) You must be around 60, either that or your systems were old at the time ;-) I started on 370s (135,138,145,148,158,168 - 3033, 4341, 3081, 3084 etc.) We wrote our assembler on punched cards and were happy to do so, And you try and tell the young people of today that . they won't believe you. Python: http://www.phespirit.info/montypython/four_yorkshiremen.htm mark -- 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
Re: Linux version
John Summerfield wrote: Gregg Levine wrote: Hello! Well for my Slackware Intel system, I have listed in there, Slackware 11.0, the command was: [EMAIL PROTECTED]:/# cat /etc/slackware-version Slackware 11.0 John is right. Of all of the distributions I've worked with, Slackware does do things differently. No. The others do things differently: I may be wrong, but I think Slackware predates the other current distros. See http://lwn.net/Distributions/ Thought wouldn't it be nice if another option were added to the uname command that listed Distributor info? It could at a minimum do the functional equivalent of cat /etc/distro-info and each distributor could set a softlink to their own existing files, or patches could be applied to uname to supply the info inline. Then customers could do something like uname -V or uname -D and have a consistent method of obtaining distributor information. mark -- 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
Re: Root filesystem
Mark Post wrote: On 8/15/2008 at 12:58 AM, in message [EMAIL PROTECTED], Mark Perry [EMAIL PROTECTED] wrote: -snip- Is there a way to trigger a script when a filesystem (FS) hits a certain percentage full? (90%?) Of course. I have such a thing set up on my Slack/390 development systems so that I know when to temporarily suspend rsynching from my upstream source at slackware.com. please elaborate ;-) If so, then one could develop a method to automatically issue the required lvresize and ext2online commands to keep the FS within a certain percentage range (70-90%?). Of course rules could be developed to make this more sophisticated: which FS are controlled, what % range per FS, limits of VG % free etc. You're certainly willing to do that to yourself. I would not want to do it, nor make that available to others. That sort of thing is very, very, complicated, and I would want a human looking at that and making decisions, not software. I am sure that view is held by others too, but for many the resource most limited is not storage, it's skilled I.T. support personnel. Any reduction in the amount of administration is usually welcomed with open arms. Others have suggested taking this one level higher to include the z/VM SMAPI to add minidisks from a pool. I think this is a more complete solution. It would allow the administrator to add disks to the z/VM pool in groups rather than ordering them individually. It would allow a much higher level company/dept. storage management. Again I would stress that any such system should be rule based, so as to prevent abuse. It has similarities to a z/OS SMS Group, where multiple z/OS LPARs within the same SYSPLEX can allocate space on-demand across a pool of shared DASD. SMS also has the option to mark pool DASD volumes as available or not, thus allowing an administrator to decide when to allow extra DASD to be made available for allocation. mark -- 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
Re: sane-dasd-update
Adam Thornton wrote: SNIP http://download.sinenomine.net/sane-dasd-update Hello Adam, I had an old(er) system which still had by-id rather than by-path so I downloaded your perl script and gave it a whirl - perfect! as shown by a diff and an ls command: ihlscob0:~ # diff /etc/fstab /etc/fstab-by-path 1c1 /dev/disk/by-id/ccw-IBM.7500029410.a840.04-part1 / ext3 data=journal,noatime,acl,user_xattr1 1 --- /dev/disk/by-path/ccw-0.0.1000-part1 /ext3 data=journal,noatime,acl,user_xattr1 1 10c10 /dev/disk/by-id/ccw-IBM.7500029410.a840.04-part2 swap swap pri=200 0 --- /dev/disk/by-path/ccw-0.0.1000-part2 swap swap pri=200 0 ihlscob0:~ # ihlscob0:~ # ls -la /dev/disk/by-path/ccw-0.0.1000-part1 lrwxrwxrwx 1 root root 12 2008-07-30 18:06 /dev/disk/by-path/ccw-0.0.1000-part1 - ../../dasda1 ihlscob0:~ # ls -la /dev/disk/by-path/ccw-0.0.1000-part2 lrwxrwxrwx 1 root root 12 2008-07-30 18:06 /dev/disk/by-path/ccw-0.0.1000-part2 - ../../dasda2 ihlscob0:~ # Thanks for a very useful tool :-) mark -- 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
Re: Striped LVM max size
Lee Stewart wrote: SNIP For example, doing an LVM with 45 mod 9s. Round numbers math gives me 6.8GB/volume x 45 volumes = 306GB. That's close to what Max gives. But what's the number with 8 stripes? Hello Lee, The idea is to put each of the specified stripes on a different DASD. I have always used a number of stripes that was a factor of the number of DASD in the VG. So for your example of 45 DASD that would mean using a number of stripes of either 9 or 5. Also I have always added extra DASD to the VG pool in multiples of the number of stripes, in the above example that would mean adding either 9 or 5 DASD at a time. If you do not do as I suggest, then you can end up with more stripes on one or more DASD than on others. This would defeat the purpose of using stripes, which is to spread the I/O load evenly across multiple DASD. If you want to use a number of stripes of 8, then either put 48 DASD in your VG pool or reduce to 40. 48 has more factors, so with 48 DASD in the VG you could choose a number of stripes from 2,3,4,6,8,12,16,24. mark -- 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
Re: Root filesystem
Mark Post wrote: On 8/14/2008 at 8:26 AM, in message [EMAIL PROTECTED], Ryan McCain [EMAIL PROTECTED] wrote: Thats the issue we are trying to avoid if possible. If we could put /, /opt, /usr, /lib, etc. etc. into LVM, we won't have to guestimate how much disk we'll need from the outset. We could grow as needed. Laid out properly, / will never grow. # df -h FilesystemSize Used Avail Use% Mounted on /dev/dasda1 388M 168M 200M 46% / /dev/mapper/vg01-home 97M 4.4M 88M 5% /home /dev/mapper/vg01-opt 74M 21M 50M 30% /opt /dev/mapper/vg01-srv 1.2G 1.1G 100M 92% /srv /dev/mapper/vg01-tmp 291M 34M 242M 13% /tmp /dev/mapper/vg01-usr 2.0G 901M 1.1G 45% /usr /dev/mapper/vg01-var 2.0G 608M 1.3G 32% /var Of course the amount of space dedicated to each LV will vary according to specific needs. The fundamental concept is the same, and will (hopefully) be the default on SLES11 if things go as I hope. Mark Post Is there a way to trigger a script when a filesystem (FS) hits a certain percentage full? (90%?) If so, then one could develop a method to automatically issue the required lvresize and ext2online commands to keep the FS within a certain percentage range (70-90%?). Of course rules could be developed to make this more sophisticated: which FS are controlled, what % range per FS, limits of VG % free etc. Hint to distros: If such a method were available during the installation process then we would not need to make guesstimates of the LV sizes, they could be set small and allowed to grow as packages were installed. If a good idea. then maybe add this to LVM2 so that the whole process was synchronized without the possibility of a FS becoming full (based on rules of course). mark -- 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
Re: vi alternative?
John Summerfield wrote: RPN01 wrote: When you get down to just 3270 access, the sed command is your friend. Do it once to the terminal, if the file isn't too big, and check your results, then use to put the results into a new file, rename the old, rename the new, and then start the cycle over again... Where possible, I like to cp file-to-change saved-file-to-change sed saved-file-to-change file-to-change \ # whatever edit commands Repeat seds until done. This approach preserves relevant permissions on the original (mv does not). you are aware that sed (and perl) has a -i command line option to create a backup? from the 3270 console I usually use perl. mark -- 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
Re: Linux sles9 felt in coma
Pieter Harder wrote: - What are your z/VM SRM parameters? The defaults aren't good for Linux workloads. I beg to disagree (sorry, Barton!). I have been running for years with the defaults and never had an E-list problem. If you do, my suggestion is that your system is wrongly set up. Most likely your paging subsystem is underpowered for the task at hand. Then you're not squeezing your hardware hard enough to get the most out of it. See next two items. - How much real and expanded storage does your z/VM system have? That's not enough, get more. Simply telling to get more real will kill the business case. At the prices IBM is charging zSeries will lose out. I know, I am up against guys running blade servers with 256GB apiece. They can afford to waste some at the price they are paying. I can't. If they don't have enough hardware to support the workload they intend to run, they need to get enough hardware. If the business case isn't there, then don't run it on System z. It's not the right tool for every job. But again, this could go back to not having the SRM parameters set appropriately for a Linux workload. If you're not over comitting storage to the extent z/VM can for that workload, you'll need more real storage than is strictly necessary. In a midsize business (like mine) it is getting to the point System z is not the right tool for any job from a business point of view. Over time that will get System z killed, as there are a lot more midsizers than really big elephants. On overcommitting storage: I have 2 pages sitting on page for each page of real I have and I am paging at hundreds per second during daytime on a 2 IFL system. I don't know I want to overcommit more. Some machines need to pull in 100K pages when coming out of low utilisation state, taking perceptible time for that . (No, no E-lists, Velocity tells me not) - Does it really need to be that big? You should try shrinking it. Some you can't shrink (eg. SAP engines, databases etc). Almost always you can, because they sized the guest according to what the midrange folks think is necessary. Very rarely have I ever run into a case where the guest was size appropriately when the site was just getting started with Linux on System z. Perhaps I should have said can't shrink any more. When machines going into low utilization (mostly dormant) get robbed of less than half their resident set over hours, and machine going idle for days go to maybe 40 percent, I don't think they can be shrunk much more, or they simply don't have the Vsize to run work at reasonable utilization. Hi Peter, as you probably know, we run several z/VM systems with many SAP Linux Guests of about 8GB each, and STORBUF defaults always cause us to have Guests on the E lists. We set STORBUF 300 300 300, and E list problems go away. After that we have options to tune with CMM and of course using VDISK. I am curious if your SAP systems are ABAP only or dual stack with JAVA. The JVM on a large SAP systems really hit storage utilization hard. mark -- 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
Re: Linux sles9 felt in coma
Pieter Harder wrote: One more thing, the ABAP stacks go dormant when no users are working, but the dual stack never goes dormant. Rob vd Heij checked for me that is due to a 10 millisec timer in Java that we can't get rid of due to SAP Java version requirements. H'mm, I wonder if the new J9 JVM will help? I think the answer to the standard STORBUF: Q1=125% Q2=105% Q3=95% not causing me grief is that I run production in Q0 due to quick dispatch being set. Production on averages owns 25-30 percent of real pages and never goes below 15 percent, so it is really hard for Q3 machines to get up to that 95 percent. QUICKDSP - well yes of course that exempts you from VM distach controls all together. That is overkill. mark -- 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
Re: swap to DCSS documentation/cookbook?
Pieter Harder wrote: Hi list, I am starting to think about the possibilities 2GB above the bar DCSS in z/VM 5.4 will provide for replacing large Vdisk swap. But I can't find any usable documentation on how to do swap to DCSS. The basics are documented on dw and I can probably manage, but when anyone has been there and done that can you please post some pointers and save us all the trouble of finding out again. Performance information is quite welcome as well. Hi Pieter, DCSS is good for sharing code (or constant data) between Guests. Swap data is unique to only one Guest. What do you hope to gain over a/multiple VDISK - size? performance? mark -- 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
Re: Linux sles9 felt in coma
Pieter Harder wrote: I think the answer to the standard STORBUF: Q1=125% Q2=105% Q3=95% not causing me grief is that I run production in Q0 due to quick dispatch being set. Production on averages owns 25-30 percent of real pages and never goes below 15 percent, so it is really hard for Q3 machines to get up to that 95 percent. QUICKDSP - well yes of course that exempts you from VM distach controls all together. That is overkill. Why overkill? It is the setup that is decided on. Give production an absolute priority above everybody else, and let the rest contend in a free-for-all. Besides, if production does go long dormant it will lose it real pages. When it becomes active again we don't want it to be held back by anything. Hi Pieter, absolute priority makes your SAP Guests compete with TCPIP and other z/VM services that are already QUICKDSP. As your SAP Guests depend on these z/VM Services to provide an efficient service your setup is not optimal. mark -- 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
Re: Linux sles9 felt in coma
Pieter Harder wrote: Why overkill? It is the setup that is decided on. Give production an absolute priority above everybody else, and let the rest contend in a free-for-all. Besides, if production does go long dormant it will lose it real pages. When it becomes active again we don't want it to be held back by anything. Hi Pieter, absolute priority makes your SAP Guests compete with TCPIP and other z/VM services that are already QUICKDSP. As your SAP Guests depend on these z/VM Services to provide an efficient service your setup is not optimal. That is new to me. What z/VM services do Linux SAP guests depend on? Certainly not the TCPIP stack machine when the guest is connected to a vswitch. The controllers maybe, but under normal circumstances they are not terribly active as most vswitch work is done by CP. I can think of the MPROUTE server being affected, but then that one doesn't have a whole lot to do in an OSPF stub area with only a default route to the rest of the world. Fair point Pieter, if one thinks of a z/VM machine dedicated to serving only Linux Guests. I have vague memories of our having run systems with all Linux Guests using QUICKDSP and it simply became a mess. I guess if you have only one Production Guest and that is your most important workload then its your call. But the more Guests you have then you need to allow z/VM to handle the (overcommitted) resources. mark -- 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
Re: update-status and parse-metadata CPU load
Mark Post wrote: On Wed, Jul 16, 2008 at 5:36 PM, in message [EMAIL PROTECTED], Ron Foster at Baldor-IS [EMAIL PROTECTED] wrote: Hello, I just returned from a disaster recovery test. After we restored our Linux for zSeries guests, we booted up several at one time. We noticed the system being very sluggish. I jumped on each of them and noticed that either a process called update-status or parse-metadata was running on all of them and consuming most if not all of the CPU. To get around the problem, I killed the offending process and went about finishing the test. chkconfig novell-zmd off You can either use YaST Online Update to do maintenance, or use the zypper command in place of rug. Fortunately, zmd is going away with SLE11, leaving only zypper and YOU. Starting with SLES10 SP2, any new installs for System z novell-zmd is disabled by default. Mark, I have SP2 systems that do not have Internet access, and on those I find that suseRegister has to be disabled too because it will start novell-zmd at boot time. chkconfig suseRegister off Mark -- 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
Re: Cobbler anyone?
Michael MacIsaac wrote: Hello list, Anyone heard of cobbler? (no, not the shoe repairman nor the deep-dish fruit dessert with a thick top crust :)) This one: http://cobbler.et.redhat.com/ I wouldn't bother asking, but it seems they are looking at a port to s390x: https://fedorahosted.org/cobbler/wiki/SssThreeNinety Just FYI... I don't see any plugin for the z/VM SMAPI ;-) mark -- 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
Re: Reiser
Rob van der Heij wrote: On Tue, Jul 15, 2008 at 6:22 AM, Alan Altmark [EMAIL PROTECTED] wrote: Not only the kernel itself, also the other commercial middleware and applications that you run there. SAP for example has a very short list of certified kernels. It's not so short, so long as its either SLES or RHEL ;-) mark -- 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
Re: Reiser
Rob van der Heij wrote: On Mon, Jul 14, 2008 at 12:49 AM, Erik N Johnson [EMAIL PROTECTED] wrote: (and I believe that is one of the reasons Red Hat is less eager to take s390 patches from IBM under the counter). What are you inferring here?, IBM patches are now all part of the mainline kernel. You can download a recent kernel from www.kernel.org and you will see s390 code is there. They may not _all_ have gone directly into 2.6.5 but they are _all_ in 2.6.25 and many have been back-ported by RedHat and Novell-Suse into RHEL 5 and SLES 10. mark -- 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
Re: Reiser
Rob van der Heij wrote: On Mon, Jul 14, 2008 at 10:04 AM, Mark Perry [EMAIL PROTECTED] wrote: Rob van der Heij wrote: On Mon, Jul 14, 2008 at 12:49 AM, Erik N Johnson [EMAIL PROTECTED] wrote: (and I believe that is one of the reasons Red Hat is less eager to take s390 patches from IBM under the counter). What are you inferring here?, IBM patches are now all part of the mainline kernel. You can download a recent kernel from www.kernel.org and you will see s390 code is there. Yes, I am aware that s390 is one of the architectures. But we do have areas where the priorities of the open source community may match those of large shops wanting to run Linux on the mainframe. I am pretty sure the SuSE distribution kernel has/had portions that were not (yet) accepted in the mainstream Linux sources (for example the instrumentation stuff for FCP chpids and the CMMA patches). I just downloaded 2.6.26 and found CMMA related code in: arch/s390/mm/page-states.c arch/s390/mm/init.c Of course this doesn't prove a time-line, but today the CMMA code is in the mainline kernel. I think the bigger problems were back-porting problems because the Enterprise releases from the Distributors are a long way behind the latest kernel releases. I just think it is unfair to imply that IBM is doing anything under the counter. If you watch the LKML you will frequently see contributions from IBM developers. mark -- 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
Re: Adding a CPU in SLES10
Mark Post wrote: On Mon, Jul 14, 2008 at 9:11 PM, in message [EMAIL PROTECTED], Jim Fujimoto [EMAIL PROTECTED] wrote: Hello List, I want to dynamically add a cpu to a single cpu SLES10 SP1 server. If this were a SLES9 server I would echo a 1 into /sys/devices/system/cpu/cpu1/online, but only the ../cpu0 directory exists on the SLES10 server. Has the method to do this changed? Yes. The kernel no longer defines 32 CPUs, whether they're available or not. Absent any kernel parms, it detects how many are present, and defines the control blocks for just that amount. If you want to have the ability to dynamically add more later, you have to use the possible_cpus=n kernel parm, where n is the maximum number you anticipate wanting. You can find more information in /usr/src/linux/Documentation/cpu-hotplug.txt Also of interest to some maybe the cpuplugd supplied on SLES-10-SP2 in s390-tools. See man cpuplugd. mark -- 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
Re: Convert existing mount point to LVM
Ceruti, Gerard G wrote: Hi All Our current disk layout has specific mount points with MOD 9's behind , we now need more space on one specific mount point, Is it possible to turn this mount point into a LVM configuration without moving the data off ?. Looking at the doco , the current plan would be to move all the data off the mount point, delete it then recreate it as LVM and then copy all the data back ?. You email is a little confusing? If I replace all of the occurrences of mount point(s) with filesystem(s) is makes a little more sense. You mention MOD-9s, so its fair to assume that you have a filesystem that has now reached 6.8GB and is maxed out. So you need to create at least 2 new PVs each of 1 by 3390-9. Add those PVs to a new VG. Create an LV of 6.8GB in the new VG. Remember to add any new DASD device numbers to your zipl.conf (make a backup first) and run zipl. Now you have a choice. Create a new filsystem in the LV and play various filesystem copy methods using cp or tar, or directly copy the whole filesystem using dd. For dd, first unmount the existing filesystem. (or at least remount it RO using mount -oremount,ro /dev/dasdx1.) (It would probably be a good idea to run e2fsck on it at this point.) Use dd to copy the /dev/dasdx1 to /dev/vgname/lvname. dd if=/dev/dasdx1 of=/dev/vgname/lvname Then run resize2fs on /dev/vgname/lvname to grow your filsystem into the new and larger LV. (Again running e2fsck on the new LV would be a good idea) Remember to update your /etc/fstab (make a backup first) with the new LV /dev/vgname/lvname in place of your old filesystem /dev/dasdx1. mount /dev/vgname/lvname should now mount it in place of the old filsystem. If the new filesystem verifies OK - you be the judge of that. Then you can recycle the old MOD-9. If there are problems then use the backup copy of your /etc/fstab to remount the old filesystem. There are many ways to skin a cat, the above is but one of them ;-) mark -- 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
EXT3+LVM2 and BARRIERS - fixed with SLES-10-SP2
I started a previous thread on Journaling Filesystems and BARRIERS in which it was discussed that EXT3 on LVM2 disables barriers. I can now report that using SLES-10-SP2 that EXT3 with LVM2 no longer disables barriers. If you value your corporate data then this is yet another reason to move to SP2 :-) mark -- 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
Re: Reiser
Erik N Johnson wrote: It may be that bug reports go unheeded for a while. The good news is it looks like 15 years and not 25 ;-) mark -- 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
mkfs.ext3 and ext2online - Info for SLES-10-SP2
Hi List, Since using SLES-10-SP2 I notice that ext3 filesystems now have new default options that better suit them for later online resizing. If you look at the changelog for e2fsprogs you will see: * Thu Dec 06 2007 - [EMAIL PROTECTED] - set EXT2_FEATURE_COMPAT_RESIZE_INODE by default in mke2fs [#279180,FATE#302553] If you do a dumpe2fs of a filesystem created on SP2 you will now see a new Filesystem feature of resize_inode, and a new line Reserved GDT blocks. I have already tested extending a 4GB ext3 filesystem beyond 16GB which use to be the first barrier I always hit. So if you want to use online resizing SP2 has advantages. Perhaps there is(or will be) an Online Update for SP1? mark -- 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
Re: DNS Entries
Tom Duerbusch wrote: Really, really dumb question. What would make a DNS server automagically add an entry into its list? If you are using DHCP and the DHCP Server is configured to update the associated DNS, that would do it automagically. Discuss with your network admin. mark -- 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
Re: NFS mount from z/os fails not owner
Ian S. Worthington wrote: Thanks Mark. I added that, but it didn't fix it. Searching around about it though, led me to the correct log file, that led me to an error message, that led me to insecure, which did. Great that you fixed it :-) An alternative would be to specify on z/OS in /etc/services a port for mountd that is below 1024. mark -- 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
Re: Zebra
Goodwin, Derric wrote: Anyone use zebra? We are trying to set it up on our z/guests for network redundancy and when trying to set the hello timer interval with Ip ospf hello-interval 3 We are getting Error occured during reading below line. hello-interval 3 Below are my conf files. Any ideas? Zebra.conf hostname quagga password quagga enable password quagga log file /var/log/quagga/quagga.log ospfd.conf hostname quagga password quagga enable password quagga log file /var/log/quagga/ospfd.log router ospf ospf router-id 10.120.40.214 ip ospf hello-interval 3 network 10.120.40.0/24 area 0 network 10.124.100.0/24 area 0 interface dummy0 interface eth0 Hello Derric, the ip ospf hello-interval 3 is an interface command, as such it should follow the interface statement to which it should apply. e.g. interface eth0 ip ospf hello-interval 3 If you use SLES10 I would recommend using the quagga rpm from the SP2 DVD , its at 0.99.9 and has a lot of bugs fixed in it. mark -- 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
Re: NFS mount from z/os fails not owner
Ian S. Worthington wrote: I'm attempting to mount a linux filesystem from z/os but am getting rc=8b/6e050001, not owner. According to the ip trace it seems the initial MOUNT is working ok but the subsequent FSINFO is failing with NFS3ERR_PERM (1) The filesystem is configured in /etc/exports as 192.168.15.0/24(ro) and I can mount it fine from itself. Any ideas what the problem might be? Hi Ian, are you using uid=0 on z/OS, and if so have you tried adding no_root_squash option to your Linux exports file? mark -- 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
Re: Adding virtual CPUs to a running Linux guest
[EMAIL PROTECTED] wrote: Folks, I need to purcahse a clue.I have a Linux guest (SLES10, SP1), running on z/VM 5.3. The directory entry for the guest has a MACHINE ESA 2 statement in it. Thye Linux guest was booted with only one virtual CPU defined. I would like to define and bring online the 2nd virtual CPU, without having to stop and reboot the Linux image. Can Linux handle the addition of an extar CPU while it is running, and, if so, how? A simple CP DEFINE CPU 1 does add an extra virtual CPU to the virtual machine configuration, but Linux does'n see it automatically Hi Dave, the running linux system needs to have already been boot with the kernel parameter: possible_cpus=number If this number was higher than 1, then yes just vary the cpu online ;-) mark -- 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
Re: Loss of network after 1st boot
Johnston, Robert E wrote: Hello Mark, I wouldn't exactly say it is a waste of time on my part. The only linux commands I know is ls! I never really know what I'm going to be working on from day to day and I wanted to get my feet wet before starting a free trial. Mark Post's recommendation of a Free-Trial is not marketing. Even from this list you will most likely get much better support if you run current versions of either SLES or RHEL. There is also a complete starter system with VM and Linux pre-wrapped which if you are a newbie makes the most sense to me. And don't worry about being a newbie, I have seen many understanding replies to newbies on this list. This list welcomes growth, so invite your friends ;-) mark -- 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
Re: src_vipa Usage
Petersen, Jim wrote: We are using quagga and ospf for z/Linux. We have a dummy0 which is the VIPA address say 1.2.3.4 We have a eth0 which is the Ethernet addresssay 10.1.1.1 We have a hsi0 which is the Hipersockets interface say 20.1.1.1 So how would one 1) code a route-map to use 1.2.3.4 for all applications whether they go out 10.1.1.1 or 20.1.1.1 And 2) how do you include the route-map in the qugga (or would it be the zebra) configuration file. Hello Jim, rather than by interface, really one would set the src-vipa based on destination. If as in your above example your destinations consist of 2 subnets 10.1.1.0/24 and 20.1.1.0/24 then a suitable route-map in /etc/quagga/zebra.conf may look similar to the following: --- password enable password xxx ip route 0.0.0.0/0 x.x.x.x route-map vipa1 permit 10 match ip address prefix-list DEST set src 1.2.3.4 continue route-map vipa1 permit 20 ip protocol ospf route-map vipa1 ip prefix-list DEST permit 10.1.1.0/24 le 32 ip prefix-list DEST permit 20.1.1.0/24 le 32 log file /var/log/quagga/quagga.log --- Not that you can then telnet into your quagga and display this list. telnet hostname 2601 show ip prefix-list detail Hope that helps ;-) mark -- 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
Re: Intrusion Detection Solution for Suse SLES 10 on zLinux
Robert Giordano wrote: I am looking for a software solution that is supported on zLinux SuSe SLES 10 SP1 for Intrusion Detection. Any thoughts is greatly appreciated! I've not really tried it, but SNORT ships with SLES See http://www.snort.org mark -- 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
Re: Unit record device driver
Mark Post wrote: On Thu, Jun 5, 2008 at 11:36 AM, in message [EMAIL PROTECTED], Pat Carroll [EMAIL PROTECTED] wrote: Hi List, I'm trying to find the latest source RPM for the ur driver. The only one I can find is ur-0.4 (from the Large Scale Linux Deployment Redbook (2002)). Anyone have a link? Anyone installed it under RH4 or 5? For the driver: http://www.ibm.com/developerworks/linux/linux390/linux-2.6.16-s390-19-october2005.html For the userspace tools: http://www.ibm.com/developerworks/linux/linux390/s390-tools-1.6.2.html Both of which are included in SLES10 SP2, but I don't know if they're in any version of RHEL yet. The vmur(.ko) is also included in RHEL 5.2 mark -- 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
Re: Unit record device driver
Mark Post wrote: On Thu, Jun 5, 2008 at 12:31 PM, in message [EMAIL PROTECTED], Mark Perry [EMAIL PROTECTED] wrote: -snip- The vmur(.ko) is also included in RHEL 5.2 How about the userspace tool(s)? look like it: ihlscob0:~ # rpm -qilp /mnt/rhel52/Server/s390utils-1.5.3-17.el5.s390x.rpm|grep vmur warning: /mnt/rhel52/Server/s390utils-1.5.3-17.el5.s390x.rpm: Header V3 DSA signature: NOKEY, key ID 37017186 /sbin/vmur /usr/share/man/man8/vmur.8.gz ihlscob0:~ # mark -- 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
Re: Unit record device driver
Pat Carroll wrote: The user space tools are there, but the driver (unless I'm missing something) is not. Oh its there: ihlscob0:~ # rpm -qilp /mnt/rhel52/Server/kernel-2.6.18-92.el5.s390x.rpm|grep vmur warning: /mnt/rhel52/Server/kernel-2.6.18-92.el5.s390x.rpm: Header V3 DSA signature: NOKEY, key ID 37017186 /lib/modules/2.6.18-92.el5/kernel/drivers/s390/char/vmur.ko ihlscob0:~ # mark -- 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
Re: Longest line in a text file?
McKown, John wrote: Is there a handy utility that I've missed which will tell me the longest line in a text file? I'm looking at something that I could use to determine if a text file should be printed in portrait or landscape mode, depending on the longest line length. Just did that today ;-) use wc -L mark -- 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
Re: lsscsi command
Ayer, Paul W wrote: Hi Anyone have the lsscsi command installed on there z/Linux systems? (I'm using RHEL 4.6 - 5.1 on s390x) I can't seem to down load it. I have seen some documents where it's in S390-tools but have not yet found it. Any pointers to where I can get it would be much appreciated. Both the Redhat 5.1 and 5.2 DVDs contain an lsscsi package: /mnt/rhel52/Server/lsscsi-0.17-3.el5.s390x.rpm /mnt/rhel51/Server/lsscsi-0.17-3.el5.s390x.rpm The contents of these rpms show an lsscsi command and a man page. mark -- 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
Re: vmcpi command
CHAPLIN, JAMES (CTR) wrote: Has anyone used the vmcp command? And reference on how to set it up on our zLinux systems. We are using RHEL 4.5 5 on zVM 5.3 LPARs. We have the man pages in place: vmcp - send commands to the z/VM control program but I get command not found when working with it. Can anyone point me to a good reference on setting this tool up? vmcp is documented in the IBm Device Drivers and Commands manual 2 parts 1) a kernel module 2) a command The kernel module comes with your kernel The module is in the 390tools package. To use modprobe vmcp vmcp q cplevel mark -- 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
Re: fstab and other small, critical file recoveries - VMUR.KO
Mike Walter wrote: Does that make sense to do? 2) What's the best way to echo/cat the critical files to the VM SPOOLed console log after every successful boot? New with SLES-10-SP2 is the VM Unit Record kernel driver vmur.ko. Some basic documentation is in the new Device Driver and Command manual. It can be used to directly interact with the RDR, PUN, PRT (default 00c, 00d, 00e). Example First ensure the vmur module is loaded either at boot or via command: modprobe vmur ihlscob2:~ # chccwdev -e c Setting device 0.0.000c online Done ihlscob2:~ # chccwdev -e d Setting device 0.0.000d online Done ihlscob2:~ # chccwdev -e e Setting device 0.0.000e online Done ihlscob2:~ # ls -la /dev/vm* crw-rw 1 root root 10, 62 2008-05-27 12:53 /dev/vmcp crw-rw 1 root root 254, 14 2008-05-30 09:13 /dev/vmprt-0.0.000e crw-rw 1 root root 254, 13 2008-05-30 09:13 /dev/vmpun-0.0.000d crw-rw 1 root root 254, 12 2008-05-30 09:13 /dev/vmrdr-0.0.000c ihlscob2:~ # vmcp q rdr ORIGINID FILE CLASS RECORDS CPY HOLD FORM DEST KEEP MSG IHLSCOB2 0016 A PUN 00089515 001 NONE STANDARD OFF ON OFF IHLSCOB2 0017 A PUN 0001 001 NONE STANDARD OFF ON OFF IHLSCOB2 0018 A PUN 00106677 001 NONE STANDARD OFF ON OFF DIRMAINT 0019 A PUN 0006 001 NONE STANDARD OFF OFF OFF ihlscob2:~ # cat /dev/vmrdr-0.0.000c rdr1 ihlscob2:~ # ls -la rdr1 -rw-r--r-- 1 root root 9404416 2008-05-30 09:15 rdr1 ihlscob2:~ # dd bs=80 if=/etc/fstab of=/dev/vmpun-0.0.000d dd: writing `/dev/vmpun-0.0.000d': Invalid argument 17+1 records in 17+0 records out 1360 bytes (1.4 kB) copied, 0.001641 seconds, 829 kB/s ihlscob2:~ # vmcp q pun ORIGINID FILE CLASS RECORDS CPY HOLD FORM DEST KEEP MSG IHLSCOB2 0020 A PUN 0017 001 NONE STANDARD OFF OFF OFF ihlscob2:~ # Of course you'll have all the issues of LRECL (see above for DD with BS=80) and EBCDIC/ASCII (use iconv). have fun ;-) mark -- 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
Re: RHEL System z Max Memory= 64GB?
Brad Hinson wrote: It's related to this kernel patch, submitted upstream a few days ago: http://git390.osdl.marist.edu/cgi-bin/gitweb.cgi?p=linux-2.6.git;a=commit;h=7f7032120414df426a9f2949caf9f6eafcea0a7e When a guest (or Linux in LPAR) has more than ~100 GB storage defined, the initrd gets overwritten when the section is cleared. I don't know if SLES or other distros have included this patch, but RHEL hasn't yet. So technically the limit is around 100 GB per z/Linux instance, but 64 GB was all we had on hand to certify at the time (last year). I contacted Heiko, he assured me that there is no such low Linux limit on System z, and (almost) 4TB is theoretically possible. However it is possible to run into the initrd problems that you mention. They can be bypassed post installation (install 64GB , change zipl to load at high address), or a permanent fix for resolving is here: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=65912a84c0f33304fa5ea004c7b6ee58d5f5572e mark -- 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
Re: RHEL System z Max Memory= 64GB?
Brad Hinson wrote: After further investigation, the limit is actually correct on the page. It's per guest though, not per system (z9/z10) so this could probably be worded better. We're investigating a couple of patches to increase this limit in RHEL 5. Where do you get a limit of 64G per guest from? VM 5.3 has no such limit? and even so, there is always LPAR mode. Is this something unique to RHEL 5.x ? mark -- 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
Re: Ext3 (Journaling Filesystems) and BARRIERS = Filesystem Integrity
Patrick Spinler wrote: How much good would a write barrier do on a linux on z/VM anyway? Would a barrier message from a linux kernel, through a minidisk and minidisk cache, through a cp device, to a disk controller, through the controller's cache preserve a barrier operation? (Please correct the i/o path as appropriate) H'mm, why not skip all of that, let's fall back to IDEs on System z ;-) Config an AOE kernel-module initiator on your zLinux and hook up some PCs with x by 1TB IDEs to your LAN switch. Make sure they run Linux and vblade target. No routing involved/allowed, no TCP stack involved. I wonder if the QETH/QDIO driver would support this? Layer2 configured spare 10GB OSA Port anybody ;-) http://en.wikipedia.org/wiki/ATA_over_Ethernet H'mm SLES-10 ships with aoe.ko mark -- 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
Ext3 (Journaling Filesystems) and BARRIERS = Filesystem Integrity
I am not sure how many of the people on this list subscribe to LWN, but there was, for me, an eye-opening article on Barriers and journaling filesystems: http://lwn.net/Articles/283161/ Even if you don't subscribe I'll try to explain the core of the document The issue is in what state is a filesystem left after a disk or Linux crash? Journaling filesystems can only be as good/safe/reliable as the integrity of the Journal file itself. If the data being written to a Journal file is written out-of-order, as can happen because of various disk write optimization techniques in both the Linux kernel and the disk hardware itself, then there is the possibility that the Journal COMMIT record get written to disk before the other Journal data records. The current solution to this is for Filesystems such as ext3 to enable barriers, which in theory should cause a sync to occur of all pending data before the COMMIT record can be written. Now here's the big catch - LVM does not currently support barriers, and so any ext3 on LVM is exposed to possible corruption. I note that RedHat defaults to putting all of their filesystems under LVM on System z right now.. The Linux Kernel defaults to not using barriers, but SLES has a patch to enable it by default. I am curious what plans there maybe, or workaround solutions, for both SLES and RHEL? mark -- 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
Re: Ext3 (Journaling Filesystems) and BARRIERS = Filesystem Integrity
Patrick Spinler wrote: One other bit of news, for what it's worth, are these: http://lkml.org/lkml/2008/2/15/126 ~ and http://lkml.org/lkml/2008/2/15/166 So, for device mapper (e.g. lvm) devices, a solution is in the works. Well in the works for 2.6.25. Now the real question is what is in the works for back-porting for SLES-10 and RHEL-5 ? Assuming that patches for dm to support barriers make it through there is still the ominous comment from Andrew Morton regarding the use of barriers and why they made the Linux default to not use barriers: Last time this came up lots of workloads slowed down by 30% so I dropped the patches in horror. I just don't think we can quietly go and slow everyone's machines down by this much... If up to 30% slowdown of a filesystem is true, then workarounds such as moving the Journal onto a separate disk start to look attractive mark -- 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
Re: Ext3 (Journaling Filesystems) and BARRIERS = Filesystem Integrity
LJ Mace wrote: What is the patch number?? thanks Mace The only reference in Patrick's email to a patch was my quoted email mention of a SUSE patch within his email. If this is what you are refering to the it is not a fix, it is in the base SLES code. Uf you run SLES you have barriers enabled by (SUSE) default. mark -- 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
Re: Ext3 (Journaling Filesystems) and BARRIERS = Filesystem Integrity
Patrick Spinler wrote: How much good would a write barrier do on a linux on z/VM anyway? Would a barrier message from a linux kernel, through a minidisk and minidisk cache, through a cp device, to a disk controller, through the controller's cache preserve a barrier operation? (Please correct the i/o path as appropriate) Better if this were answered by a z/VM and Linux developer, and hardware development, and., but my 5-cents z/VM MDCs are write-through and hence a completed write operation means the data has been accepted by the DASD subsystem before a response is sent back to Linux. The DASD subsystems have options such as fast-write, which means they would tell z/VM the write was complete before the data hits a real disk. Of course such systems usually come with Non-Volatile memory for their caches and hence claim data integrity across DASD subsystem restarts. But your question really is the one that any one interested in the integrity of their company's data should be asking ;-) mark -- 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
Re: chccwdev_hotplug_udev troubles in SLES9_SP4 ?
Kurt Verhofstadt wrote: Hi, We recently started migrating our SLES9_SP2 s390x-servers to SLES9_SP4. We have been using udev, mainly for switching our 3590 and 3592 tapedrives between the VM guests. The linux-user 'db2inst1' does the db2backup to tape. This user sends the mount request to VM and after completion, does 'chccwdev -e 0.0.0cuu' to put the tapedrive online. Since SP4, this resulted in an error because this user has no write-rights on /sys/devices/css0/path/channel-id/online. Our udev.rules are setup so that a tape-device name /dev/tape/cuu is created. On this device, the user can write. But, udev is only called after chccwdev -e is executed. And this does no longer work. We bypassed this by adding 'chown' and 'chmod' in /etc/hotplug/ccw.agent to allow a group (with db2inst1) gets write access to '/sys/devices/css0/path/channel-id/online'. Are there any other users that came across this problem ? Is it (still) possible for non-root users to execute 'chccwdev' ? Thanks for any tips or advise. Kurt is it not better to do the chccwdev via sudo and authorise db2inst1 to the sudo list? See man sudo and man sudoers for more info. mark -- 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
Re: AW: How to add VIPA permanently
Brad Hinson wrote: Hi Anna, Do you have a link to the Redbook you used to setup VIPA? I'd like to run through the steps on RHEL 4 to see what the process is to make the changes persistent across IPLs. Hello Brad Anna, I am not yet aware of Anna's exact VIPA requirements. But just a heads-up. IBM's z/OS and z/VM have VIPA keywords for their TCPIP PROFILEs, but on Linux the concept of a VIPA is vague. If using a dummy device for the VIPA, e.g. dummy0, then at boot time the dummy kernel module must be loaded. If dummy0 is used then by default the OSA's will not respond to arp requests for the new VIPA until the VIPA is registered with the OSA card. I typically use an if-up POST-UP script that runs qethconf vipa add x.x.x.x ethx (from the s390 tools) for each OSA interface. Alternatives to using a dummy device include such things as adding a secondary IP address to an existing interface (even lo). But historically this has led to problems configuring things such as OSPF via zebra/quagga. In any case I usually set the netmask to 255.255.255.255 (/32) because you do not want to create the impression that a genuine subnet (of VIPAs) exists. Again this can cause confusion with OSPF which may advertise a route to the VIPA subnet via another host which also utilises a VIPA. On z/OS a VIPA is understood to be a HOST and OMPROUTE will advertise the VIPA as a HOST address (this is configurable too.) -- Mark Perry Director RITA Co., Ltd. Uttaradit, 53000 Thailand email: ritacoltd (at) googlemail (dot) 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
Re: RHEL System z Max Memory= 64GB?
Mark Perry wrote: Can anyone from RedHat comment as to why their webpage: http://www.redhat.com/rhel/compare/ Shows System z has having a maximum amount of memory of 64GB? A z10 can have 1.5 TB, and I have heard of Linux tests above 512GB. I note that one point is the RHEL 5.2 announcement was support for 512GB (not system specific). So a potential web browsing customer may simply dismiss System z as being limited to only 64GB.. mark -- 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
Re: SLES10 and NFS problems.
Hi Bernhard, on the NFS Server system have you tried using the command: service nfsserver restart Also rather than relying on yast why not simply edit /etc/exports ? Regarding the clients hanging - this depends on the nfs mount option hard or soft. The whole idea with hard is that an nfs server can stop and restart (or network outages can occur) without causing data loss to the client. The nfs client is made to wait until the server responds. If you switch to a soft mount option, then a permanent i/o error will be report via the nfs client to you application(s). mark Bernard Wu wrote: Hi List, I have 2 SLES10 guests. In both guests NFS4_SUPPORT is set to no in /etc/sysconfig/nfs . I then ran SuSEconfig on both guests. Now the problems. 1. On the NFS server, after adding files to be exported ( using yast2 ) , I get Unable to restart the NFS server. Your changes will be active after a reboot . How can I activitate the changes without going through a reboot ? 2. On the client side, I have to issue the command reautofs start before the exported files can be seen. Did automount change between SLES9 and SLES10. 3. Although I can see the exported files , I cannot access those files. If I try to copy one of those exported files to /tmp, the session hangs. Nothing seems to be able to kill the cp. Not even a kill -9 . Once the session hangs, the exported files cannot be seen. The only way to clear up the mess seems to be a reboot. Help. The information contained in this e-mail message is intended only for the personal and confidential use of the recipient(s) named above. This message may be an attorney-client communication and/or work product and as such is privileged and confidential. If the reader of this message is not the intended recipient or an agent responsible for delivering it to the intended recipient, you are hereby notified that you have received this document in error and that any review, dissemination, distribution, or copying of this message is strictly prohibited. If you have received this communication in error, please notify us immediately by e-mail, and delete the original message. -- 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
Re: Ruby on Rails
David Boyes wrote: (OK, ANSI C trigraphs were a really bad idea, but c'est la vie. Every language has at least one obligatory wart.) Trigraphs saved my bacon, oh um 20+ years ago. C development on OS/2 (was that Watcom or Zortech C?) downloaded to MVS compiled with SAS C. trigraphs saved some complex codepage conversion problems. We mixed SAS C with IBM Assembler (only way to do real multitasking at that time) Ah, the good old days.. But yes, orrible! mark -- 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
Re: Ruby on Rails
Mark Perry wrote: Trigraphs saved my bacon, oh um 20+ years ago. sorry gross exaggeration, must have been 1991, so only 17 years ;-) mark -- 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
Re: How to add VIPA permanently
Fuhrmann Anna wrote: Hi all, [ the List sent me a mail rejecting my post to this matter letting me know that I submitted the message more than once - I did not do it as far as I know, anyway I submit it once more as I did not find any answers, sorry for insisting ] The VIPA question: I defined VIPA im sys file system so that it can be activated without IPL, there is a cookbook explaining how to do that. But I don't know how to make the changes permanent - i.e.: how does VIPA get started after IPL? Does it also need an ifcfg-dummy0 in /etc/sysconfig/network-scripts? And if yes, how about the parameters: what is DEVICE - dummy? what is NETTYPE - qeth or something else? what is TYPE - Ethernet or something else? what about SUBCHANNELS - probably not necessary? Anyone around using VIPA? Or knowing where to find documentation as to defining it in zLinux? Hello Anna, I would have liked to have given you a definitive answer with examples, but we do not test RHEL 4. It is possible that I may be able to produce some VIPA examples on RHEL 5.1/5.2 if I can find some spare time to do so. The majority of our HA testing is done on SLES 10 (SP1 and SP2). Let me know if examples from these other Linux versions/distros is of any use to you? mark -- 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
Re: AW: How to add VIPA permanently
Fuhrmann Anna wrote: Hi Mark, I would have liked to have given you a definitive answer with examples, but we do not test RHEL 4. It is possible that I may be able to produce some VIPA examples on RHEL 5.1/5.2 if I can find some spare time to do so. The majority of our HA testing is done on SLES 10 (SP1 and SP2). Let me know if examples from these other Linux versions/distros is of any use to you? I can't really tell you if it would be of any use as I don't know SUSE Linux i.e. how far SUSE configuration for zLinux is the same as on RHEL ... thanks Hello Anna, SUSE and RedHat to be blunt, differ. They have lots in common, but things like configuration commands/scripts are the major areas of difference. Sure there are similarities of course, but I don't want to give you something and have it not work on RHEL 4 which I do not run. We only test the new stuff here ;-) Even on my PC I run RHEL 5.1 ;-) mark -- 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
RHEL System z Max Memory= 64GB?
Can anyone from RedHat comment as to why their webpage: http://www.redhat.com/rhel/compare/ Shows System z has having a maximum amount of memory of 64GB? A z10 can have 1.5 TB, and I have heard of Linux tests above 512GB. mark -- 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
Re: z/Linux access to z/OS DASD
McKown, John wrote: Very correct. Ah now I understand the buggy problems with so many programs that were certified as correct. They haven't yet reached the very correct level yet ;-) Hey we can take mathematics to a whole new level, why stop at infinite, we can now have very infinite ;-) mark -- 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
