Hi Som, The experiment I suggested is to prove that the boot-archive has already corrupt before you reboot the machine and problem exists in your write handler.
IIRC, boot-archive use mkfile(1M) to create a file and use the file as a lofi device. Since the size of boot-archive is not correct, I suspect the command mkfile failed during create archive. You can uncompress a correct archive from CD and check the size of the lofi device before compress. Then I think you can try mkfile(1M) command like "mkfile $size file1" on your iscsi target device with your initiator driver so that you can simplify the scenario and debug/trace the problem during write operation. Is it possible you handle (DMA) incorrectly when the buffer size is big enough? IE. (multiple cookies or have to bind partially and split window to transfer). Generally, we don't use analyze function of format to verify our driver. We use diskomizer or vdbench which are really good test utilities to test a HBA driver. Cheers Javen Somnath kotur wrote: >Javen/Juergen, > Thank you for the tip,yes i have tried the same >below ... mounted it on /a ,and checked the archive by >typing the command below: > ># gunzip < boot_archive > /tmp/bootarchive.img > >resulted in another error saying >'gunzip:stdin: invalid compressed data - format >violated' > >So it did not really help ,tho the size of the >bootarchive seemed to reduce from 38M to 20M > >I then used the 'analyze' option in the 'format' >utility of solaris. This in turn has an option called >'verify' that writes the entire disk and verifies the >contents in many passes > >Ran tests for few hrs and they all passed!! > >The only other problem i could think of was that >whenever i attempted to create a filesystem on my LUN >using: > >newfs -f /dev/rdsk/c1t1d0s2 > >i get an error saying: > > 'mkfs: close failed on write disk: I/O error .' > > Although i get this error im always able to >succesfully mount and read /write files from the >volume ,so i decided to ignore this > >However when i did an 'fsck' on the above special file >i did seem to get lot of inode errors ,and fsck is >taking time repairing all of them. > >I got a lot of errors even after doing step 1 below >,identifying my BOOT volume and then doing fsck on it > >Is there any other utility or some option that you can >suggest to identify any write errors ? > >Thanks >Som > > > >--- Javen Wu <[EMAIL PROTECTED]> wrote: > > > >>FYI >> >> >>>Date: Mon, 03 Mar 2008 02:38:00 +0000 >>> >>> >>From: Javen Wu <[EMAIL PROTECTED]> >>To: Somnath kotur <[EMAIL PROTECTED]> >>CC: [EMAIL PROTECTED] >>Subject: Re: Fwd: iSCSI LUN Boot >> >>Hi Som, >> >>My point of view, the problem could not caused by >>Synchronize_cache failure. >>Because synchronize_cache is not a mandatory command >>in SCSI spec, that >>means >>even without the command, the system can work fine. >> >>My guess there is some error on your handler for >>WRITE. In another >>words, I guess >>something wrong during your driver write buffer out. >> >>Could u do a experiment as below: >>1. boot the machine with Solaris CD and your ITU >>disk >>2. using `bootadm update-archive -R $ROOT` to update >>the boot-archive on >>your iSCSI target. >>Here $ROOT should be your mount point of your >>alternative root. >>3. before reboot, could you verify the new >>boot-archive under your >>$ROOT/platform/i86pc/ corrupt or not. >>4. if the image correct, please reboot the machine >>from your iSCSI >>target again. >> >>If #3 is failed, that means your write has some >>problem. So please do >>test write and verify with your driver. >> >>Cheers >>Javen >> >> >> >> >>>I am attempting to do an iSCSI LUN boot using my >>> >>> >>iSCSI >> >> >>>HBA driver with the 8/10/07 Solaris CD. Able to >>> >>> >>feed >> >> >>>my driver using the ITU diskette with my driver >>>package succesfully, my driver gets detected ,LUN >>> >>> >>for >> >> >>>boot is made available for selection and the entire >>>installation for the selected distribution (tried >>> >>> >>many >> >> >>>options from entire to just 'reduced network >>> >>> >>support') >> >> >>>goes through 100% ,eventually the 'reboot' screen >>>prompt appearing . >>> >>>I'm copy pasting >>> >>> >>> >>> >>>>below >>>> >>>> >>>> >>>> >>>>>the first few lines of the install_log and also >>>>> >>>>> >>>>> >>>>> >>>>the end of the file >>>> >>>> >>>> >>>> >>>>> >>>>> >>>>> >>>>> >>/*****************************************************/ >> >> >>> >>> >>> >>> >>>>>Installation partially failed >>>>>Error: 1332 blocks >>>>>Error: >>>>>Error: Processing package instance <MYIscsi> from >>>>></tmp/19039> >>>>>Error: >>>>>Error: My Dummy iSCSI HBA Driver for >>>>> >>>>> >>>>> >>>>> >>>>Solaris(i86pc) >>>> >>>> >>>> >>>> >>>>>1.0.0 >>>>>Error: Using </a> as the package base directory. >>>>>Error: ## Processing package information. >>>>>Error: ## Processing system information. >>>>>Error: 3 package pathnames are already >>>>> >>>>> >>properly >> >> >>>>>installed. >>>>>Error: ## Verifying disk space requirements. >>>>>Error: >>>>>Error: Installing My Dummy iSCSI HBA Driver for >>>>>Solaris as <MYIscsi> >>>>>Error: >>>>>Error: ## Installing part 1 of 1. >>>>>Error: /a/kernel/drv/amd64/myiscsi >>>>>Error: /a/kernel/drv/myiscsi >>>>>Error: [ verifying class <none> ] >>>>>Error: ## Executing postinstall script. >>>>>Error: >>>>>Error: Installation of <MYIscsi> was successful. >>>>>Error: >>>>>Error: *** IMPORTANT NOTICE *** >>>>>Error: This machine must now be rebooted >>>>> >>>>> >>>>> >>>>> >>>>in >>>> >>>> >>>> >>>> >>>>>order to ensure >>>>>Error: sane operation. Execute >>>>>Error: shutdown -y -i6 -g0 >>>>>Error: and wait for the "Console Login:" >>>>>prompt. >>>>> >>>>> >>>>> >>>>> >>>>> >>>/************ END OF THE FILE >>> >>> >>> >>> >>>>***********************/ >>>> >>>> >>>> >>>> >>>>>Installing boot information >>>>> - Updating boot environment configuration >>>>> >>>>> >>>>> >>>>> >>>>file >>>> >>>> >>>> >>>> >>>>> - Installing boot blocks (c1t0d0) >>>>> - Installing boot blocks >>>>> >>>>> >>>>> >>>>> >>>>(/dev/rdsk/c1t0d0s0) >>>> >>>> >>>> >>>> >>>>>svc.configd: Fatal error: >>>>> >>>>> >>>>> >>>>> >>>>/a/etc/svc/repository.db: >>>> >>>> >>>> >>>> >>>>>integrity check failed. Details in >>>>>/etc/svc/volatile/db_errors >>>>>svccfg: Repository server failed (exit 102). >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> >>>/**************************************************/ >>> >>> >>>> >>>> >>>> >>>> >>>On reboot, i get GRUB error saying 'Error >>> >>> >>28:selected >> >> >>>item cannot fit into memory' >>> >>>I was able to boot into the system back using CD >>> >>> >>and >> >> >>>my driver,mount the FS of my LUN and using gunzip >>>verified that my bootarchive is corrupt >>> >>>One thing that i constantly get in my system log >>>through the installation is a warning msg from the >>>SCSI driver saying >>>scsi: SYNCHRONIZE CACHE COMMAND FAILED(5) >>> >>>Couldnt find much help on this and if this is the >>>rootcause of the problem >>> >>>BTW,i tried local install of the OS on a SCSI HD >>> >>> >>using >> >> >>>Adaptec SCSI HBA and it worked fine >>> >>>PS: I have also tried installing with 'kd' enabled >>> >>> >>and >> >> >>>kmem_flags set to check kernel heap memory >>> >>> >>corruption >> >> >>>and nothing showed up >>> >>>Pls advise >>> >>>Thanks >>>Som >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >>> >____________________________________________________________________________________ > > >>>Looking for last minute shopping deals? >>>Find them fast with Yahoo! Search. >>> >>> >http://tools.search.yahoo.com/newsearch/category.php?category=shopping > > >>> >>> >>> >>> >=== message truncated ===> >_______________________________________________ > > >>driver-discuss mailing list >>[email protected] >> >> >> >http://mail.opensolaris.org/mailman/listinfo/driver-discuss > > > > > > > ____________________________________________________________________________________ >Never miss a thing. Make Yahoo your home page. >http://www.yahoo.com/r/hs > > _______________________________________________ driver-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/driver-discuss
