Bug 654323 (trainframe)
Summary: | Failure on Create file-system from the FBA Edev device during installation formatting Failure | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Product: | Red Hat Enterprise Linux 6 | Reporter: | Tong <huynh_tong> | ||||||||
Component: | anaconda | Assignee: | Samantha N. Bueno <sbueno> | ||||||||
Status: | CLOSED CURRENTRELEASE | QA Contact: | Release Test Team <release-test-team-automation> | ||||||||
Severity: | high | Docs Contact: | |||||||||
Priority: | low | ||||||||||
Version: | 6.0 | CC: | brueckner, dcantrell, dhorak, hamzy, harihare, jstodola, mgrf, nasastry, ssant, tlavigne | ||||||||
Target Milestone: | rc | Keywords: | Reopened | ||||||||
Target Release: | --- | ||||||||||
Hardware: | s390x | ||||||||||
OS: | Linux | ||||||||||
Whiteboard: | |||||||||||
Fixed In Version: | Doc Type: | Bug Fix | |||||||||
Doc Text: | Story Points: | --- | |||||||||
Clone Of: | |||||||||||
: | 708145 (view as bug list) | Environment: | |||||||||
Last Closed: | 2013-09-23 19:28:13 UTC | Type: | --- | ||||||||
Regression: | --- | Mount Type: | --- | ||||||||
Documentation: | --- | CRM: | |||||||||
Verified Versions: | Category: | --- | |||||||||
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||||
Cloudforms Team: | --- | Target Upstream Version: | |||||||||
Embargoed: | |||||||||||
Bug Depends On: | 708145 | ||||||||||
Bug Blocks: | 835616, 885957, 960065 | ||||||||||
Attachments: |
|
Filesystem package contains just basic system directory layout. It has nothing to do with installation or filesystems generally. As the error occured in installation procedure, reassigning to Anaconda for further investigation. Hi Ondrej, Is there any update on this problem, Thanks, No update yet. I try to see if problem still there, and let you know. regards, Tong FYI... Just re-installation RHEL6.0 with Edev the problem still the same. New and extended product functionality as well as new features for this component must be implemented by IBM and accepted upstream before consideration for a RHEL backport can be made. For your convenience, this bug has been cloned to the same component under Fedora, which serves as the upstream development area for RHEL. Development Management has reviewed and declined this request. You may appeal this decision by reopening this request. Created attachment 687345 [details] rhel63-based updates.img I'm attaching an updates.img based on RHEL6.3, as requested. I have not actually been able to replicate this bug, despite repeated effort. However, after discussing this with some colleagues and doing some reading, I do think I found the problem, so hopefully this fix should work. To test: 01) upload the updates.img to some web space 02) pass: updates=http://url.to/updates.img when booting the installer If this does not work, please attach all log files. If a traceback occurs, you need only attach the anaconda-tb file. Thanks. Sachin, Hariharan, as said - put you on cc: for testing the update.img Even the installation on normal dasd also fails with panic on both LPAR and z/VM. log : dracut: Starting plymouth daemon dracut: rd_NO_DM: removing DM RAID activation dracut: rd_NO_MD: removing MD RAID activation dracut Warning: No root device "block:/dev/mapper/vg_r3515051-lv_root" found dracut Warning: LVM vg_r3515051/lv_swap not found dracut Warning: LVM vg_r3515051/lv_root not found dracut Warning: Boot has failed. To debug this issue add "rdshell" to the kernel command line. dracut Warning: Signal caught! dracut Warning: LVM vg_r3515051/lv_swap not found dracut Warning: LVM vg_r3515051/lv_root not found dracut Warning: Boot has failed. To debug this issue add "rdshell" to the kernel command line. Kernel panic - not syncing: Attempted to kill init! CPU: 0 Not tainted 2.6.32-279.el6.s390x #1 Process init (pid: 1, task: 000000003f67b120, ksp: 000000003f67fc20) 000000003f67fd70 000000003f67fcf0 0000000000000002 0000000000000000 000000003f67fd90 000000003f67fd08 000000003f67fd08 00000000004d7244 000000003fe412ef 000000008001590c 0000000080000000 000000000080c000 000000000000000d 000000000000000c 000000003f67fd60 0000000000000000 0000000000000000 00000000001055ec 000000003f67fcf0 000000003f67fd30 Call Trace: ([<00000000001054ec>] show_trace+0xe8/0x138) [<00000000004d707e>] 01: HCPGSP2629I The virtual machine is placed in CP mode d ue to a SIGP stop from CPU 01. panic+0xda/0x24c [<00000000001501fa>] do_exit+0x80a/0x968 [<00000000001503b6>] do_group_exit+0x5e/0xec [<0000000000150478>] SyS_exit_group+0x34/0x40 [<0000000000118f12>] sysc_noemu+0x10/0x16 [<000003fffd4a73ba>] 0x3fffd4a73ba 00: HCPGIR450W CP entered; disabled wait PSW 00020001 80000000 00000000 00112590 I have gone through the 'updates.img' - this doesn't contain the 'parted binaries parted, partprobe and library libparted.so'. So, the functionality of EAV and EDEV won't work. My basic understanding is 'updates.img' will contain any Anaconda related changes and 'install.img' will contain all the required binaries and libraries. So we should get 'install.img' with the required parted binaries and libraries. updates.img contents: # gunzip -dc ../updates.img | cpio -id 1303 blocks # ls storage # cd storage/ # ls dasd.py deviceaction.py devicelibs devices.py devicetree.py errors.py fcoe.py formats __init__.py iscsi.py miscutils.py partitioning.py partspec.py storage_log.py udev.py zfcp.py # ls devicelibs/ crypto.py dm.py edd.py __init__.py lvm.py mdraid.py mpath.py swap.py # ls formats/ disklabel.py dmraid.py fs.py __init__.py luks.py lvmpv.py mdraid.py multipath.py prepboot.py swap.py To confirm tried installation with EAV, EDEV devices and the installation FAILED. Created attachment 692606 [details]
rhel63 updates.img
The updates.img can and should now contain the parted changes. The new libparted should unpack to /tmp/updates and should be getting loaded. The new parted binary should unpack to /tmp/updates/sbin; if you want to call it directly, you'll need to adjust your shell's PATH or LD_LIBRARY_PATH.
I apologize for not incorporating the parted changes earlier earlier; try out this new updates.img and let me know if that works for you.
Hello Samantha, Here are the tests performed based on the reworked update.img you attached 2-04 In addition to the known customer scenario 8. Installation on EDEV device .... item 8. in the list below .... ... the following test scenarios got performed based on the update.img to support EDEV and EAV. 1. EAV full disk installation with full package 2. EAV disk, with minimal partition size and minimal package installation 3. Installation on LVM on EAV 4. Installation on RAID on EAV 5. Installation on DASD device 6. Installation on SCSI device 7. Installation on Multipath device 8. Installation on EDEV device ( including scsi emulated FBA device ) is completed . 9. Installation on Mini disk 10. Upgrade using normal disk 11. Full package installation on RAID device 12. Installation over cms formatted disk So from the above test coverage the update.img should be ready to be verified on customer site for installation on EDEV devices. The following limitations got observed that should not impact the customer 1) When repeatedly use same disk for installation, auto partition fails to delete the existing partition workaround: Through custom partition its possible to create partitions Fix tracking for R 6.5 with 87262 - RH885755 - Auto partition fails to delete existing partition 2) During test execution sometimes installation on EAV Disk using LVM layout failed - BZ will be opened when root cause identified and fix for distribution is required 3) Upgrade from 6.3 to 6.4 fails for FBA device. (No option is displayed for upgrade in anaconda) I prepare a summary to cleanup all bugs related to this topic - Fedora and RHEL - not simple Question: Am I right that all the patches that got delivered with the update.img are also included in Fedora 18? This bug got used to track patch inclusion for the update.img That is delivered meanwhile -> so propose to close this bugzilla The remaining work to include the fixes into Fedora 19 will be tracked via separate BZs ... will be listed her as soon we have agreement on further processing (In reply to comment #18) > I prepare a summary to cleanup all bugs related to this topic - Fedora and > RHEL - not simple > > Question: > Am I right that all the patches that got delivered with the update.img are > also included in Fedora 18? The parted changes are in F18, but the anaconda changes are not. (In reply to comment #19) > This bug got used to track patch inclusion for the update.img > > That is delivered meanwhile -> so propose to close this bugzilla Oh, also--I'd like to leave this bug open. It's flagged and approved for 6.5, and even aside from that, I never received any confirmation from the customer that the updates.img from comment 15 was suitable (need to ping appropriate parties again). This request was evaluated by Red Hat Product Management for inclusion in a Red Hat Enterprise Linux release. Product Management has requested further review of this request by Red Hat Engineering, for potential inclusion in a Red Hat Enterprise Linux release for currently deployed products. This request is not yet committed for inclusion in a release. Patches posted for review. Based on bug 631928 comment 94, it looks as though we now have functional support for this. We originally anticipated changes required for parted _and_ anaconda which prompted the separate bugs for each component being opened; however, it seems only changes to parted were necessary in order to gain support for this. That said, I'm closing this bug since no changes to anaconda were ultimately needed. Thanks to everybody who helped out. |
Created attachment 461090 [details] Bug_file_out_from_installation_script Description of problem: z/VM Version 6 Release 1.0, service level 1002 (64-bit) Generated at 10/22/10 15:38:04 EST running as guest under zVM6.1 using the FBA Edev device and get and failure during the installation procedure. Error: An error was encounter while formatting device /dev/dasddb2 details: invalid device specification Version-Release number of selected component (if applicable): Component FCP (FBA edev device) RHEL6.0 GA Linux version 2.6.32-71.el6.s390x (mockbuild.bos.redhat.com) (gcc version 4.4.4 20100726 (Red Hat 4.4.4-13) (GCC) ) # 1 SMP Wed Sep 1 01:38:33 EDT 2010 How reproducible: set edev e04a type fba attr scsi fcp_dev f200 wwpn 50604082d52e6982 lun 004a000000000000 vary on e04a cpfmtxa e04a Steps to Reproduce: running the installation PUN RH6GA KERNEL PUN RHEL PARM PUN RH6GA INITRD CH RDR ALL KEEP NOHOLD IPL 00C following the installation procedure when you get to creating the file system and formatting that where the failure occur. Error: An error was encounter while formatting device /dev/dasddb2 details: invalid device specification Actual results: Expected results: Additional info: Please see the attachment file that generate from the installation procedure.