SlackwareThis Forum is for the discussion of Slackware Linux.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
Have finally retired my old Pentium II box (running Slack 10 from the command line) and got a 'new' box: P4-ht, 3.0Ghz, 1G ram, DVD-ro (ide) and support for SATA HDD. The 'new' box is a few years old and came w/o HDD disk. Thought I'd drop in a new HDD, install Slack 12.2 and enjoy Slack with graphics and speed.
The Slack installation fails w a segmentation fault. I ran a bios ram check....it passed all tests.
More detail: Installed a 500GB HDD SATAII and the bios sees it. Downloaded Slack to CDs and it booted from the DVD drive. Created 4 /dev/sda partitions of 10G, 240G, 238G and 2G for /, /home, /usr and SWAP, respectively. The formating goes ok (reiserfs on all partitions). The failure hits while packages are copying from disk 1. I tweaked several bios settings but the failure seems to halt at the same spot each time (on kernal-modules).
The HDD was Seagate and on-line conversation said the latest firmware (SD15) has a boot problem. The firmware update I downloaded, scanned and saw the HDD, but then said it didn't see the HDD when attempting to install the update. Thinking such inconsistency pointed to a bad HDD, I returned the Seagate in favor of a 500G Western Dig. HDD. Installed the HDD (bios sees it), but the install w the WD HDD failed....at the same spot during the install as the seg fault using the Seagate. It's not the HDD.
Thought the install CDs were bad. Checked MD5...all looked OK, but still didn't trust them. Downloaded and burned the DVD iso. DVD install went exactly the same as with the CDs....with a failed seg fault during the kernel-module copy. It's not the install media.
Skimmed LQ for other threads...found one which suggested this might be a ram problem. Although the bios ram check reported no issues, I'm thinking I may be swapping out the ram next.
I am a newbie (w limited budget) and thought I'd tap into the gurus here before pulling the ram. Any other thoughts on what may be happening?
while your in bios set them to non windows or non-OS if you use the rieserfs then make sure you are selecting the huge kernel smp. well to tell you the truth it still could be the dvd burn and for some reason you keep burning it and it keeps failing. there is a bad package but that does not cause a segfault. mmmmmm I will tell you I have seen a lot of people installing the reiserfs FS have had problems.could you do a small test with a ext3 fs. give it a try. what do you have to loose
H_TexMex_H: I used the full install (vs advanced) so I didn't have the option of selecting what gets installed. But your suggestion is good in that I'll also try an advanced install and see if I can further narrow down where the fault occurs.
Thanx for the other suggestions and links. I'll do a bit more probing and will post the results.
The age of the equipment and possible lack of maintenance would indicate that you should first pull the memory. You could have a dirty edge on the stick. You could try cleaning the system and connector(s).
You should first use good ground techniques, either with a grounding wrist strip or at least placing yourself at the the same potential level as the system by maintaining contact with the equipment.
I would shut the power off to the system and unplug the system. Clean the card edge for the memory using strips of new clean 20# paper with denatured alcohol. Swipe in one direction with folded wet paper for the edge, a new piece for each swipe. You can use a credit card to clean the connector if that is needed by folding the paper over the end of the credit card. But first you should use dry air to blow the connector out. In the USA it can be purchased at most computer/office supply houses, even Walmart has it. Then replace the card without touching the card edge.
BTW, don't use a eraser to clean/buffer the card edge. You are doing more harm then good.
I would then run 'memtest86' for a number of passes, possibly over night. If you get errors then you have a problem. Replace the bad stick. If the memory is used with bad data/address problems then you will getting nothing but nightmares!
I would not rely on the memory test performed by the 'POST'. 'memtest86' performs several pattern tests that will exercise the cells. There are better test but 'memtest has become a standard plus it's free.
onebuck: Thanks for your suggestion re cleaning and memory checks.
The problem is solved!
Details: Poking around further in the bios revealed a setting for the type of boot. Setting was Quickboot (QB) versus Fullboot (FB). Doing a FB performed a full memory test....a test not performed w QB. I set to FB and rebooted. Surprise, Surprise, the memory test revealed bad ram. To be sure, I popped the ram modules (the edge fingers looked clean) and swapped them. The problem followed the module. Good news.
Ran off to the store, bought new Kingston 2G ram DIMMs for a steal (on sale for US$40 ea), dropped them in and completed the Slack install w/o a hitch. It was very straightforward! (I'm on the new install as I type this!)
I don't know why the bios memory check ran previously didn't catch it. It reported no issues. Oh well.
Thanks again to all for the LQ help....what a terrific resource!
Most 'POST' memory test don't perform a full pattern test. Takes too long!
If you run memtest you will notice the pattern test that are performed. Not just a walking one pattern but different full pattern tests.
Other memory test utilities can run a stress test an perform timing an duration tests. But most of these are not open source ($$). Some memory test beds perform with embedded software that can really stress the stick but they are very expensive for individual purchase.
Glad to see you got the system going. As for the edge and connector. I would perform the cleaning as suggested. You might get surprised. The edge can look clean but if you ever look at one under a scope you would be surprised. We would clean our system memory and all edges then apply a contact protector, one trade name that comes to mind is 'Chromilin R'. That too is expensive for private use. The last time I purchased a oz it was over $100 but it will last for a long time if you use it properly.
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.