Bug 17993
Summary: | Redhat 7.0 Installation failed at the postinstall tasks receiving signal 11 | ||
---|---|---|---|
Product: | [Retired] Red Hat Linux | Reporter: | Vladimir Gendler <vgendler> |
Component: | kernel | Assignee: | Arjan van de Ven <arjanv> |
Status: | CLOSED CURRENTRELEASE | QA Contact: | Brock Organ <borgan> |
Severity: | high | Docs Contact: | |
Priority: | high | ||
Version: | 7.0 | CC: | chabotc, juanjo, vgendler |
Target Milestone: | --- | ||
Target Release: | --- | ||
Hardware: | i386 | ||
OS: | Linux | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | Bug Fix | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2004-09-30 15:38:50 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: |
Description
Vladimir Gendler
2000-10-01 04:46:15 UTC
That normally indicates out of memory (obviously not in your case since you have 384Mb) or bad memory. Has this box been running Linux before ? Reassigned to kernel - doesnt feel like an installer error. Its too random I agree that it is the kernel problem. In addition to this RAM I defined 3 swap partitions, so I have plenty of memory. I doubt that my memory is bad. First, I have (Corsair brand) ECC memory. Second, I am running four operating systems (W98, WNT, OS2, LINUX) on this machine using the boot manager. This machine is used very heavily and I have not got any memory problem so far. Third, the previous version of RedHat - 6.2 is working perfectly. I found one thing which causes this problem. It is the EMACS package, the installation failed on this step (big package EMACS) several times. When I removed it from install I managed to install the system. But there is another problem. I was trying to install Oracle8i (v 8.1.6) for LINUX which I had on the RH 6.2. It failed on the database creation step. When it goes to the second task - creating database files it stays there forever and does not create any. I reinstall RH 7.0 several times (without EMACS) and tried to install Oracle. This problem is consistent. Eventually I installed the previous RH 6.2 and installed this Oracle without any problem as I did before. I think something wrong with glibs as well in addition to the kernel. Addtional configuration info - LINUX partition is more than 3 GB and as I mention above 3 swap partitions, 133 MB each. I have had the same bug, on several types of systems including: ----------------------------------- amd k7 800 workstation 512 Mb ram adaptec 2940 / IBM DDSY 36.8Gb HD 3Com 3c905 NVidia GeForce 2 DDR 64Mb Creative emu10k1 Plextor 40x scsi Plextor 12x/8x/4x scsi writer teac 4x/12x writer (ran 6.2 / windows before) ---------------------------------- Celeron 400 (Dell inspiron 7500 notebook) 64Mb ram 6 gig HD Neomagic vga ide cd/dvd ---------------------------------- Dual p3 - 600 (server) 512 Mb ram soft-raid0 - 240 gig (4x70) scsi 18Gb seagate baracuda (/ filesystem) 3com 3c905b NVidia GeForce (1) and all systems, i either ran into these problems, or errors while formatting partitions (see my submited bug report on this [Bug 18320]) If it is the kernel problem how do you explain that RH 6.2 is working with the same kernel? I think we have two problems here. One is the installer problem because the installation behave pretty consisyently - stops at one of the EMACS package. Another is probably glibs because Oracle does not work under 7.0 an works under 6.2 with the same kernel. Because some of the Oracle componenets get linked we can suspect glibs. I've had the same (or similar) problem. I tried many times to perform the Workstation install of RH7.0 on my system and most of the time it halted on signal 11. It did install successfully a couple of times, but never consecutively. Also, the signal 11 always happened at some point during the package installations. It would happen sometimes after only a few packages had been installed and sometimes right at the end. During the install process I noticed that occassionally the system would pause, leaving the HDD-busy light on, for up to 15 seconds or so. After this pause, it would either continue or receive a signal 11 (although, there was not always a pause before receiving the signal 11). The drive configuration during the failed attempts was: CD-ROM - secondary master, the two hard drives - primary master and slave. In this configuration win98 and ME install with no problems. The only way I was able to get RH7.0 to install consistently (three times in a row) was to slave my CD-ROM drive to one of my hard drives. Any ideas as to why this works?? The system: EPOX EP-3VBM motherboard (9-13-2000 BIOS date) PC Power & Cooling Standard 145 SFX power supply 128MB PC100 SDRAM (Crucial Technology, CT16M72S4D8E) Celeron 366 w/128KB Cache (FPO/BATCH #:L9140496) 2 Western Digital Ultra ATA/66 20GB hard drives 1 Creative Blaster 52X CD-ROM drive Riva TNT AGP graphics card Netgear FA310TX ethernet card Panasonic PanaSync S17 monitor Microsoft Natural Keyboard Fellowes 'Browser 2' ps/2 mouse By switching over to a shell before beginning the package installation process of the installation, I was able to run the hdparm utility (had to copy it from a floppy) to set certain options for the two hard drives (-u1 -d1 -c3 -m8 -X68, I think). With my two hard drives on the primary channel and my CD-ROM drive on the secondary, I've installed RH7.0 several times using this method and never received signal 11. I hope this helps... It occours with RH 7.1 too, i was running RH6.2 in my box and want to upgrade to RH7.1, but it is impossible. Thanks for the bug report. However, Red Hat no longer maintains this version of the product. Please upgrade to the latest version and open a new bug if the problem persists. The Fedora Legacy project (http://fedoralegacy.org/) maintains some older releases, and if you believe this bug is interesting to them, please report the problem in the bug tracker at: http://bugzilla.fedora.us/ |