Update: I let the box sit at the "stuck point" overnight. When I looked at it this morning, I have a prompt back with a message:
The local database is locked. It is possible that a database journal file exists that needs to be rolled back, but you don't have write permission to the database. I'm logged in as root, so how can not have permission? Figuring something was wrong, I attempted to conary rollback, but I didn't use the command correctly so it didn't run. Thinking, "What the hell?", I ran conary updateall again and this time it seemed to finished "correctly" (without the huge delay) after applying 17 update jobs. I say it that way because I don't know what "correct" looks like, but I ended up at a shell prompt with no glaring errors on screen. I've tried hitting the web page again, but it's still blank. So I rebooted. Before I could get to the keyboard, I noticed 3 choices in the boot loader. I wasn't able to read them before the bootloader timed out and started the first image in the list. After a reboot, the web page is still blank. I tried a conary updateall to make sure everything was cool, but said there were no new troves. So, I would expect at this point to have working box, but I clearly don't. I have no idea what to do now. Irebooted the box and looked at the boot choices this time. The first choice does seem to be the correct one, with a newer kernal and non SMP. Still a blank web page. <slight rant> I know Openfiler is free, so don't get on my case about what I should expect for nothing; this comes from a programmer that takes pride in his work doing exactly what it's supposed to do before anyone else ever sees it. I thought the point of the Openfiler project was to produce a burnable disk image that could be used to turn an ordinary lump of computer equipment into a fully functional network attached storage appliance. When I think of "network appliance" I don't think of a shell prompt, I think of web page. I don't think of Linux, I think of "I don't care what it runs". When I think "fully functional", I think it should work as advertised after installation, maybe with the need to run an update and probably make lots of configuration changes (all using a web page). Since beginning testing with v1 versions and now v2.2, on my current hardware and on other hardware, I've never seen an Openfiler instance that I would consider beyond a Beta version, let alone v2.2. I would never consider using this software in production or paying for support given the problems I've seen myself and the problems I've read about in the mailing lists and forums. I want the product to work and I don't mind paying for support, but the product first has to be at least beyond beta. If I want beta software, I'll use MS products. Is my understanding of the OF project not correct? Am I being too critical? Should I just drop it and find something else? I've even considered rolling my own (with just the functionality I need, iSCSI), which I'm fairly confident I could do in an amount of time comparable to the time I've wasted so far. </slight rant> Anyway, can anyone suggest what I should do to get this system functional? -Tab On Fri, March 16, 2007 12:37 am, Tab Bennedum wrote: > I tested an older v1 version of OF a while ago and have been > lurking ever since. Now that 2.2 is out, I'm back to testing again. > And I'm > disappointed. > > My system is brand new hardware: Athlon 64 running on an MSI > nforce4 board (K9NBPM2-FID, v5.3 BIOS), 1 GB RAM, 80GB Seagate PATA > system disk and 320GB Seagate SATA data disk. > > I've installed the OF 2.2 i86_64 ISO image. Immediately after > install, I try to hit the web page (http://<box>:446/) and I'm > greeted by a blank page. A view source shows an empty HTML doc > (HTML/BODY begin/end tags, no content). Hmm. Is the ISO that > broken? > > I should note that I receieved no errors (that I could identify) > during installation. > > Ok, so I ssh in as root and run "conary updateall". I get this: > > > Applying update job 1 of 19: > Install > kernel(:runtime)=2.6.19.7-0.1-1[~!kernel.debug,~!kernel.debugdata,~! > kernel.numa,~kernel.smp] Applying update job 2 of 19: > Install > kernel(:runtime)=2.6.19.7-0.1-1[~!kernel.debug,~!kernel.debugdata,~! > kernel.numa,~!kernel.smp] Creating database transaction (1 of 2)... > > > And conary is hung. I can suspend the process and kill it, but > before doing so, I left it in the above state for over an hour with > no progress being made. > > I started from scratch again and reinstalled, this time > partitioning as described in the docs (I did it a little differently > the first time). Long story short, I get the same results. > > Anyone have any clue what might be going on? What other information > can I provide that might help? Didn't find anything in the forums > that applied, and been reading the mailing list for a couple > months. Anyone? > > > I'm getting ready to show this system to someone (tomorrow) who > might be interested in putting together an iSCSI SAN but as it is, it > will never see the light of day and some other solution will be > used. > > -Tab > > > > _______________________________________________ > Openfiler-users mailing list > [email protected] > https://lists.openfiler.com/mailman/listinfo/openfiler-users > > > > _______________________________________________ Openfiler-users mailing list [email protected] https://lists.openfiler.com/mailman/listinfo/openfiler-users
