Re: zLinux support of non-mainframe disk storage
On Thu, Aug 16, 2012 at 03:51:29PM -0600, Mark Post wrote:
> >>> On 8/16/2012 at 05:11 PM, David Boyes wrote:
> > Does it still need a EW segment, or is it now well behaved for shared R/O?
>
> I guess it does. From /usr/src/linux/arch/s390/kernel/early.c:
> hlen = snprintf(defsys_cmd, DEFSYS_CMD_SIZE,
> "DEFSYS %s 0-%.5X EW %.5X-%.5X SR %.5X-%.5X",
> kernel_nss_name, stext_pfn - 1, stext_pfn,
> eshared_pfn - 1, eshared_pfn, end_pfn);
>
> #ifdef CONFIG_BLK_DEV_INITRD
> if (INITRD_START && INITRD_SIZE) {
> sinitrd_pfn = PFN_DOWN(__pa(INITRD_START));
> einitrd_pfn = PFN_UP(__pa(INITRD_START + INITRD_SIZE));
> min_size = einitrd_pfn << 2;
> hlen += snprintf(defsys_cmd + hlen, DEFSYS_CMD_SIZE - hlen,
> " EW %.5X-%.5X", sinitrd_pfn, einitrd_pfn);
> }
> #endif
>
> snprintf(defsys_cmd + hlen, DEFSYS_CMD_SIZE - hlen,
> " EW MINSIZE=%.7iK PARMREGS=0-13", min_size);
> defsys_cmd[DEFSYS_CMD_SIZE - 1] = '\0';
> snprintf(savesys_cmd, SAVESYS_CMD_SIZE, "SAVESYS %s \n IPL %s",
> kernel_nss_name, kernel_nss_name);
> savesys_cmd[SAVESYS_CMD_SIZE - 1] = '\0';
>
> __cpcmd(defsys_cmd, NULL, 0, &response);
>
>
> I can't say I understand why EW is being used. Perhaps one of the guys in
> IBM development can comment on that.
EW is being used for the first MB, since it contains the lowcore
page 0 and 1 and therefore can't be shared.
In addition we need EW for the RW data section of the kernel and
- if present - for the initrd since it gets freed after initial
usage.
Shared are the text section and the RO data section of the kernel.
--
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: zLinux guest multiprocessors
>Yes, it is the application. If it doesn't have highly parallel >multi-threading such that it can keep two CPUs working, then the guest >ends up doing extra work to deal with two processors without any benefit. Hello Alan, Thanks a lot!! We can control linux's virtual processor to horizontal and vertical polarization. What situation we should use vertical polarization? And what situation we should use the other polarization? Best Regards! Gao Lu (高路) I/T Specialist Global Technology Services IBM Global Services (China) Company Limited. Address:18/F, Pangu Plaza, No.27, Central North 4th Ring Road, Chaoyang District, Beijing, 100101 地址：北京市朝阳区北四环中路27号盘古大观写字楼18层，100101 BeiJing 100027, PRC Cell Phone: 15001327619 Internet ID: lu...@cn.ibm.com
Re: zLinux guest multiprocessors
On Tuesday, 08/21/2012 at 11:08 EDT, Lu GL Gao wrote: > Based on my understanding, machine guests such as linux on z/VM can get cpu > resource by their SHARE value. > For example, LINUXA system has SHARE RELATIVE 100. LINUXB system also has > SHARE RELATIVE 100. So LINUXA and LINUXB both has 50% of real physical > processor resource. > If LINUXA has 2 virtual processors defined in DIRECTORY, and LINUXB has > only 1 virtual processor. What difference between LINUXA and LINUXB? > > Question1: my understanding is following, is it correct? > LINUXA and LINUXB both has 50% of real cpu resource. Because LINUXA has 2 > virtual processors, every virtual processor of it has 25% of real cpu > resource. > > Question2: As far as cpu resource is concerned, what difference between > LINUXA and LINUXB? For the answer to these two questions, please see the responses to your same questions vis a vis z/OS over on IBMVM. > Question3: usually what factor determines defining multiple virtual > processor for linux guest? Application determines? Yes, it is the application. If it doesn't have highly parallel multi-threading such that it can keep two CPUs working, then the guest ends up doing extra work to deal with two processors without any benefit. Alan Altmark Senior Managing z/VM and Linux Consultant IBM System Lab Services and Training ibm.com/systems/services/labservices office: 607.429.3323 mobile; 607.321.7556 alan_altm...@us.ibm.com IBM Endicott -- 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/
zLinux guest multiprocessors
Based on my understanding, machine guests such as linux on z/VM can get cpu resource by their SHARE value. For example, LINUXA system has SHARE RELATIVE 100. LINUXB system also has SHARE RELATIVE 100. So LINUXA and LINUXB both has 50% of real physical processor resource. If LINUXA has 2 virtual processors defined in DIRECTORY, and LINUXB has only 1 virtual processor. What difference between LINUXA and LINUXB? Question1: my understanding is following, is it correct? LINUXA and LINUXB both has 50% of real cpu resource. Because LINUXA has 2 virtual processors, every virtual processor of it has 25% of real cpu resource. Question2: As far as cpu resource is concerned, what difference between LINUXA and LINUXB? Question3: usually what factor determines defining multiple virtual processor for linux guest? Application determines? Best Regards! Gao Lu (高路) I/T Specialist Global Technology Services IBM Global Services (China) Company Limited. Address:18/F, Pangu Plaza, No.27, Central North 4th Ring Road, Chaoyang District, Beijing, 100101 地址：北京市朝阳区北四环中路27号盘古大观写字楼18层，100101 BeiJing 100027, PRC Cell Phone: 15001327619 Internet ID: lu...@cn.ibm.com
Re: subversion on SLES 11
Dominic, > it was on the first DVD. Thanks - I'll stop after the first DVD download... "Mike MacIsaac" -- 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: subversion on SLES 11
Hi, On x86_64, it was on the first DVD. I downloaded the iso from this link and it seems that there is only 2 DVD for each platform : http://download.novell.com/Download?buildid=NgW3ToaagDQ~ Regards. On Aug 21, 2012 9:06 PM, "Michael MacIsaac" wrote: > > It's on the SDK. > Which one? I see three: > > -rw-r- 1 root suse 2094888960 Feb 16 2012 > SLE-11-SP2-SDK-DVD-s390 > x-GM-DVD1.iso > -rw-r- 1 root suse 4726601728 Feb 16 2012 > SLE-11-SP2-SDK-DVD-s390 > x-GM-DVD2.iso > -rw-r- 1 root suse 2224711680 Feb 16 2012 > SLE-11-SP2-SDK-DVD-s390 > x-GM-DVD3.iso > > > "Mike MacIsaac" > > -- > For LINUX-390 subscribe / signoff / archive access instructions, > send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or > visit > http://www.marist.edu/htbin/wlvindex?LINUX-390 > -- > For more information on Linux on System z, visit > http://wiki.linuxvm.org/ > -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For more information on Linux on System z, visit http://wiki.linuxvm.org/
Re: subversion on SLES 11
> It's on the SDK. Which one? I see three: -rw-r- 1 root suse 2094888960 Feb 16 2012 SLE-11-SP2-SDK-DVD-s390 x-GM-DVD1.iso -rw-r- 1 root suse 4726601728 Feb 16 2012 SLE-11-SP2-SDK-DVD-s390 x-GM-DVD2.iso -rw-r- 1 root suse 2224711680 Feb 16 2012 SLE-11-SP2-SDK-DVD-s390 x-GM-DVD3.iso "Mike MacIsaac" -- 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: subversion on SLES 11
>>> On 8/21/2012 at 08:47 PM, Michael MacIsaac wrote: > Does SLES 11 SP2 have a subversion RPM? It's on the SDK. Mark Post -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For more information on Linux on System z, visit http://wiki.linuxvm.org/
Re: subversion on SLES 11
Hi, It does. If I remember correctly, subversion is located on the SDK DVD. Regards. On Aug 21, 2012 8:48 PM, "Michael MacIsaac" wrote: > Does SLES 11 SP2 have a subversion RPM? > > "Mike MacIsaac" > > -- > For LINUX-390 subscribe / signoff / archive access instructions, > send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or > visit > http://www.marist.edu/htbin/wlvindex?LINUX-390 > -- > For more information on Linux on System z, visit > http://wiki.linuxvm.org/ > -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For more information on Linux on System z, visit http://wiki.linuxvm.org/
subversion on SLES 11
Does SLES 11 SP2 have a subversion RPM? "Mike MacIsaac" -- 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: RHEL 6.2 will not IPL
One thing I ran into on RHEL6: Ext4 seems to be a bit more sensitive about being properly closed than ext3 did, and RHEL6 by default doesn't go to a shell prompt when install is done (at least when using kickstart). When I built a test system with ext4 and a default kickstart setup, the system would build properly, but the kickstart hung at the end after the "post" script, so I was simply doing a CP IPL to start the built guest. (RHEL5 drops you into a shell prompt where you could issue a reboot command.) IPL'ing without a clean shutdown produced a number of nasty file system errors and cross links. I've since added the "shutdown" option to my kickstart file, and that corrected it. You might have ended up with a corrupted selinux configuration and incomplete policies that made it behave this way. You can also force a clean shutdown by signaling from another VM instance. -Original Message- From: Linux on 390 Port [mailto:LINUX-390@vm.marist.edu] On Behalf Of Michael MacIsaac Sent: Tuesday, August 21, 2012 2:25 PM To: LINUX-390@vm.marist.edu Subject: Re: [LINUX-390] RHEL 6.2 will not IPL Hi, > I'll follow up with more info on the can't login/SE Linux issue later. I reinstalled RHEL 6.2, this time dasdfmt'ing the two disks from a root shell before using the "install shell". I turned off SE Linux by setting SELINUX=permissive in /etc/selinux/config, and ran through the same steps including adding the many RPMs. Now my system seems much happier. So I actually changed two things, but feel that first formatting the DASD from the install root shell is what made the difference. Thanks for your help. "Mike MacIsaac" -- 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/ -- This message w/attachments (message) is intended solely for the use of the intended recipient(s) and may contain information that is privileged, confidential or proprietary. If you are not an intended recipient, please notify the sender, and then please delete and destroy all copies and attachments, and be advised that any review or dissemination of, or the taking of any action in reliance on, the information contained in or attached to this message is prohibited. Unless specifically indicated, this message is not an offer to sell or a solicitation of any investment products or other financial product or service, an official confirmation of any transaction, or an official statement of Sender. Subject to applicable law, Sender may intercept, monitor, review and retain e-communications (EC) traveling through its networks/systems and may produce any such EC to regulators, law enforcement, in litigation and as required by law. The laws of the country of each sender/recipient may impact the handling of EC, and EC may be archived, supervised and produced in countries other than the country in which you are located. This message cannot be guaranteed to be secure or free of errors or viruses. References to "Sender" are references to any subsidiary of Bank of America Corporation. Securities and Insurance Products: * Are Not FDIC Insured * Are Not Bank Guaranteed * May Lose Value * Are Not a Bank Deposit * Are Not a Condition to Any Banking Service or Activity * Are Not Insured by Any Federal Government Agency. Attachments that are part of this EC may have additional important disclosures and disclaimers, which you should read. This message is subject to terms available at the following link: http://www.bankofamerica.com/emaildisclaimer. By messaging with Sender you consent to the foregoing. -- 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: RHEL 6.2 will not IPL
Hi, > I'll follow up with more info on the can't login/SE Linux issue later. I reinstalled RHEL 6.2, this time dasdfmt'ing the two disks from a root shell before using the "install shell". I turned off SE Linux by setting SELINUX=permissive in /etc/selinux/config, and ran through the same steps including adding the many RPMs. Now my system seems much happier. So I actually changed two things, but feel that first formatting the DASD from the install root shell is what made the difference. Thanks for your help. "Mike MacIsaac" -- 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: RHEL 6.2 will not IPL
On Tuesday, 08/21/2012 at 09:39 EDT, Filipe Miranda wrote: > I think you best shot for now is to start RHEL in init 1 or Single user mode, > and put SELinux in permissive mode. > If it works, probably RACF is not behaving well with SELinux. Linux and RACF have no interaction unless Linux is talking to a RACF-backed LDAP server. Alan Altmark Senior Managing z/VM and Linux Consultant IBM System Lab Services and Training ibm.com/systems/services/labservices office: 607.429.3323 mobile; 607.321.7556 alan_altm...@us.ibm.com IBM Endicott -- 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: RHEL 6.2 will not IPL
Sam, > That disk might still be attached to the other guest and you did not detach > them, so when you ipled your guest the system's disk was mounted read only? No, I had them R/W. Thanks for the append though... "Mike MacIsaac" -- 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: RHEL 6.2 will not IPL
> Remounting root filesystem in read-write mode That disk might still be attached to the other guest and you did not detach them, so when you ipled your guest the system's disk was mounted read only? Sam Bass -Original Message- From: Linux on 390 Port [mailto:LINUX-390@VM.MARIST.EDU] On Behalf Of Michael MacIsaac Sent: Tuesday, August 21, 2012 9:02 AM To: LINUX-390@VM.MARIST.EDU Subject: Re: RHEL 6.2 will not IPL Filipe, > I think you best shot for now is to start RHEL in init 1 or > Single user mode, and put SELinux in permissive mode. > If it works, probably RACF is not behaving well with SELinux. Thanks for the quick reply (and welcome to the land of z :)) I did not try run level 1. What I did try was logging off, DETACHing the DASD from SYSTEM, ATTACHing it to another Linux (two actually, both a SLES and a RHEL) and trying to mount the root file system. The other systems were not able to mount /dev/dasd1. Curious - so maybe SE Linux was an effect and not the cause. I do not have time to debug, so I commented out the two DASD, grabbed two more DASD for minidisks 100 and 101 and updated the user directory entry. I reinstalled 6.2 onto the new disks *remembering to first start a root shell and dasdfmt the disks* (that is something I forgot to do the first time and had to do later after the first intall failed). Let me digress and say that this is an issue we've been trying to get fixed for years. It is mentioned on p 119 of the RHEL 6 Cookbook ( http://www.redbooks.ibm.com/abstracts/sg247932.html) with the following words: Important: If the minidisks 100-102 (dasdb-dasdd) have not been formatted for Linux by dasdfmt before this installation, you should see the window shown in Figure 7-5 on page 118. However, there is a known issue in RHEL 6 where this window does not open and you do not have the ability to format the disks through the installer. If you proceed without formatting the disks with dasdfmt, the installation process will fail later. If this is the case, perform the following steps: ... There was a bugzilla opened for this in RHEL 6, but I don't have the number handy. Brad knows about it and almost had it fixed in the Anaconda python code. If you can re-open it at Red Hat, that would be most appreciated. (sorry, don't mean to dump on you so soon :)) I'll follow up with more info on the can't login/SE Linux issue later. I did see two 3390-9s in /proc/dasd/devices listed as both 7042 and 7043 MB. I'm now wondering if forgetting to dasdfmt the first time is the cause of this problem. Again, welcome to system z - it will be good to have a Red Hat rep on the list again. "Mike MacIsaac" -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For more information on Linux on System z, visit http://wiki.linuxvm.org/ -- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -- For more information on Linux on System z, visit http://wiki.linuxvm.org/
Re: RHEL 6.2 will not IPL
Filipe, > I think you best shot for now is to start RHEL in init 1 or > Single user mode, and put SELinux in permissive mode. > If it works, probably RACF is not behaving well with SELinux. Thanks for the quick reply (and welcome to the land of z :)) I did not try run level 1. What I did try was logging off, DETACHing the DASD from SYSTEM, ATTACHing it to another Linux (two actually, both a SLES and a RHEL) and trying to mount the root file system. The other systems were not able to mount /dev/dasd1. Curious - so maybe SE Linux was an effect and not the cause. I do not have time to debug, so I commented out the two DASD, grabbed two more DASD for minidisks 100 and 101 and updated the user directory entry. I reinstalled 6.2 onto the new disks *remembering to first start a root shell and dasdfmt the disks* (that is something I forgot to do the first time and had to do later after the first intall failed). Let me digress and say that this is an issue we've been trying to get fixed for years. It is mentioned on p 119 of the RHEL 6 Cookbook ( http://www.redbooks.ibm.com/abstracts/sg247932.html) with the following words: Important: If the minidisks 100-102 (dasdb-dasdd) have not been formatted for Linux by dasdfmt before this installation, you should see the window shown in Figure 7-5 on page 118. However, there is a known issue in RHEL 6 where this window does not open and you do not have the ability to format the disks through the installer. If you proceed without formatting the disks with dasdfmt, the installation process will fail later. If this is the case, perform the following steps: ... There was a bugzilla opened for this in RHEL 6, but I don't have the number handy. Brad knows about it and almost had it fixed in the Anaconda python code. If you can re-open it at Red Hat, that would be most appreciated. (sorry, don't mean to dump on you so soon :)) I'll follow up with more info on the can't login/SE Linux issue later. I did see two 3390-9s in /proc/dasd/devices listed as both 7042 and 7043 MB. I'm now wondering if forgetting to dasdfmt the first time is the cause of this problem. Again, welcome to system z - it will be good to have a Red Hat rep on the list again. "Mike MacIsaac" -- 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: RHEL 6.2 will not IPL
Michael,
I think you best shot for now is to start RHEL in init 1 or Single user mode,
and put SELinux in permissive mode.
If it works, probably RACF is not behaving well with SELinux.
If you need help with that, let me know,
Regards,
Filipe Miranda
Linux on System z
On 21/08/2012, at 09:33 AM, Michael MacIsaac wrote:
> Hi,
>
> I have a RHEL 6.2 system that I logged into OK yesterday and did some work
> (including installing a number of RPMs). This morning I was unable to
> login through SSH, though I'm sure I used the right password.
>
> So I went to the console which still had root logged on and tried the
> "passwd" command - I got an error about a lacking shared library.
>
> I made the mistake of reIPLing, now can't even login from the console :((
> Here are the pertinent boot messages:
>
> IPL 100
> ... # all normal looking boot messages - then:
> Checking all file systems.
> Ý/sbin/fsck.ext4 (1) -- /¨ fsck.ext4 -a /dev/dasda1
> /dev/dasda1: clean, 5230/32832 files, 46268/131072 blocks
> Ý/sbin/fsck.ext4 (1) -- /opt¨ fsck.ext4 -a /dev/mapper/system_vg-opt_lv
> /dev/mapper/system_vg-opt_lv: recovering journal
> /dev/mapper/system_vg-opt_lv: clean, 11/65536 files, 6190/65536 blocks
> Ý/sbin/fsck.ext4 (1) -- /tmp¨ fsck.ext4 -a /dev/mapper/system_vg-tmp_lv
> /dev/mapper/system_vg-tmp_lv: clean, 13/65536 files, 6192/65536 blocks
> Ý/sbin/fsck.ext4 (1) -- /usr¨ fsck.ext4 -a /dev/mapper/system_vg-usr_lv
> /dev/mapper/system_vg-usr_lv: clean, 40793/131072 files, 290969/524288
> blocks
> Ý/sbin/fsck.ext4 (1) -- /var¨ fsck.ext4 -a /dev/mapper/system_vg-var_lv
> /dev/mapper/system_vg-var_lv: clean, 1695/98304 files, 21102/98304 blocks
> Ý OK ¨
> Remounting root filesystem in read-write mode: Ý OK ¨
> type=1400 audit(1345548478.898:6): avc: denied { add_name } for pid=962
> comm=
> "mount" name="mtab~962" scontext=system_u:system_r:mount_t:s0
> tcontext=unconfine
> d_u:object_r:admin_home_t:s0 tclass=dir
> type=1400 audit(1345548478.898:7): avc: denied { add_name } for pid=963
> comm=
> "mount" name="mtab~963" scontext=system_u:system_r:mount_t:s0
> tcontext=unconfine
> d_u:object_r:admin_home_t:s0 tclass=dir
> type=1400 audit(1345548478.898:8): avc: denied { add_name } for pid=964
> comm=
> "mount" name="mtab~964" scontext=system_u:system_r:mount_t:s0
> tcontext=unconfine
> d_u:object_r:admin_home_t:s0 tclass=dir
> type=1400 audit(1345548478.898:9): avc: denied { add_name } for pid=965
> comm=
> "mount" name="mtab~965" scontext=system_u:system_r:mount_t:s0
> tcontext=unconfine
> d_u:object_r:admin_home_t:s0 tclass=dir
> type=1400 audit(1345548478.908:10): avc: denied { add_name } for pid=967
> comm
> ="mount" name="mtab~967" scontext=system_u:system_r:mount_t:s0
> tcontext=unconfin
> ed_u:object_r:admin_home_t:s0 tclass=dir
> EXT4-fs (dm-3): mounted filesystem with ordered data mode. Opts:
> can't create lock file /etc/mtab~962: Permission denied (use -n flag to
> override
> )
> ... then things go downhill pretty fast
>
> So the "avc: denied" messages smell of SELinux. Has anyone seen this?
> I've never had this problem with RHEL 6.2 before. Perhaps pertinent is
> that this system has RACF customized as in section 18.3 of the new
> Cookbook. Has anyone had issues with RACF and RHEL 6.2/SE Linux
> interaction? Thanks.
>
> "Mike MacIsaac"
>
> --
> 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/
RHEL 6.2 will not IPL
Hi,
I have a RHEL 6.2 system that I logged into OK yesterday and did some work
(including installing a number of RPMs). This morning I was unable to
login through SSH, though I'm sure I used the right password.
So I went to the console which still had root logged on and tried the
"passwd" command - I got an error about a lacking shared library.
I made the mistake of reIPLing, now can't even login from the console :((
Here are the pertinent boot messages:
IPL 100
... # all normal looking boot messages - then:
Checking all file systems.
Ý/sbin/fsck.ext4 (1) -- /¨ fsck.ext4 -a /dev/dasda1
/dev/dasda1: clean, 5230/32832 files, 46268/131072 blocks
Ý/sbin/fsck.ext4 (1) -- /opt¨ fsck.ext4 -a /dev/mapper/system_vg-opt_lv
/dev/mapper/system_vg-opt_lv: recovering journal
/dev/mapper/system_vg-opt_lv: clean, 11/65536 files, 6190/65536 blocks
Ý/sbin/fsck.ext4 (1) -- /tmp¨ fsck.ext4 -a /dev/mapper/system_vg-tmp_lv
/dev/mapper/system_vg-tmp_lv: clean, 13/65536 files, 6192/65536 blocks
Ý/sbin/fsck.ext4 (1) -- /usr¨ fsck.ext4 -a /dev/mapper/system_vg-usr_lv
/dev/mapper/system_vg-usr_lv: clean, 40793/131072 files, 290969/524288
blocks
Ý/sbin/fsck.ext4 (1) -- /var¨ fsck.ext4 -a /dev/mapper/system_vg-var_lv
/dev/mapper/system_vg-var_lv: clean, 1695/98304 files, 21102/98304 blocks
Ý OK ¨
Remounting root filesystem in read-write mode: Ý OK ¨
type=1400 audit(1345548478.898:6): avc: denied { add_name } for pid=962
comm=
"mount" name="mtab~962" scontext=system_u:system_r:mount_t:s0
tcontext=unconfine
d_u:object_r:admin_home_t:s0 tclass=dir
type=1400 audit(1345548478.898:7): avc: denied { add_name } for pid=963
comm=
"mount" name="mtab~963" scontext=system_u:system_r:mount_t:s0
tcontext=unconfine
d_u:object_r:admin_home_t:s0 tclass=dir
type=1400 audit(1345548478.898:8): avc: denied { add_name } for pid=964
comm=
"mount" name="mtab~964" scontext=system_u:system_r:mount_t:s0
tcontext=unconfine
d_u:object_r:admin_home_t:s0 tclass=dir
type=1400 audit(1345548478.898:9): avc: denied { add_name } for pid=965
comm=
"mount" name="mtab~965" scontext=system_u:system_r:mount_t:s0
tcontext=unconfine
d_u:object_r:admin_home_t:s0 tclass=dir
type=1400 audit(1345548478.908:10): avc: denied { add_name } for pid=967
comm
="mount" name="mtab~967" scontext=system_u:system_r:mount_t:s0
tcontext=unconfin
ed_u:object_r:admin_home_t:s0 tclass=dir
EXT4-fs (dm-3): mounted filesystem with ordered data mode. Opts:
can't create lock file /etc/mtab~962: Permission denied (use -n flag to
override
)
... then things go downhill pretty fast
So the "avc: denied" messages smell of SELinux. Has anyone seen this?
I've never had this problem with RHEL 6.2 before. Perhaps pertinent is
that this system has RACF customized as in section 18.3 of the new
Cookbook. Has anyone had issues with RACF and RHEL 6.2/SE Linux
interaction? Thanks.
"Mike MacIsaac"
--
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/
