=== asac_ is now known as asac [07:07] hello [07:07] bala: Good day. [07:07] thnx [07:07] anyone has any idea how to port ubuntu mobile on a menlow board???? [07:08] Does the menlow image not work for you? [07:08] i just downloaded a USB image of ubuntu hardy for menlow [07:08] from cd image site [07:09] i tried creating two partitons in USB [07:10] both ext3 and used dd tool to write image onto the pendrive when i power up the board it hangs up [07:10] could u plz help me with some documentation on porting ubuntu on menlow board/platforms [07:10] Hmm. I'm not sure you want the partitions. I believe the image expects to create a bootable VFAT filesystem on the entire key. [07:11] allright [07:11] then [07:11] where will the image install??? [07:11] i just haave another USB driver where it can load the OS [07:11] drive* [07:12] I believe it installs on the first discovered SCSI disk (which might be SCSI, SATA, PATA, USB, or 1394). [07:12] actually what happens is that it at times searches for the devices [07:12] And it can't find any devices? [07:12] but finds none even though i have a pen drive connected [07:13] any problems with the fs on the pen drive??? [07:13] I'm not sure: I don't have any devices without built-in storage, so haven't tried that. [07:14] any documentation u can find to help me out???? [07:15] Not really. The images are created with MIC: you might look at fiddling with the chroot to see if you can get it to work. Does the device boot off an i386 livecd? [07:15] (or image thereof) [07:16] actually the board that i use is shipped from intel it doesnt have any CD drives/HDDs [07:16] no storage present [07:16] Right. Do you not have a bootable USB optical drive for testing? [07:16] Alternately, you can put the CD image on a USB key to make sure that works. [07:17] ok [07:17] Essentially, the point wouldbe to determine if the device discovery was a problem with the kernel or the installer. [07:17] but m using the DD command to write the .img file to USB [07:17] thats a correct step that m following isnt it? [07:17] If it's the kernel, it likely needs deeper fiddling than if it's just that the presented device names don't match that which the installer is expecting. [07:18] ok [07:18] Yes, for the images at the URL in the topic, dd is the right way to copy to the USB key. Many people have reported better succes rates using bs=1024k [07:19] ok [07:19] sometimes it says cant find root VFS error [07:20] That makes me think it's a problem booting the USB device: it may be in an unexpected place. You might try fiddling with the arguments from the grub menu. [07:21] yeah but when i connect the USB it states boot: there is no option there reg the grub menu just have to press enter to select the vmlinuz kernel [07:23] bala: Try typing menu at the grub prompt. [07:23] there is no grub prompt [07:28] I thought the boot warning on the USB key was a grub prompt. Hmm. [07:29] Hold on a bit, and I'll get a test environment up. [07:29] ok [07:36] RIght. I've just booted off the mccaslin image (I don't have any menlow devices), and I get a "boot:" prompt, at which I am told to press enter. [07:36] yeah exactly [07:36] That boot: prompt is a grub prompt. [07:36] ohh [07:36] ok actually m just starting off with this so did nt know [07:36] ok [07:37] so u mean to say in grub i shd modify the setiings so that the target installation device is USB? [07:38] It sounds to me like you're encountering booting issues, rather than installation issues. Am I missing something? [07:39] not booting the USB device in wich i have just the image boots if i have a vfat fs on it not if two partitions but once it does it doesnt find the target device which is again USB (2nd ) to install into [07:41] Oh, you are trying to install to another partition on the same device? I thought you were trying to install to a different USB device. [07:41] yes a diff USB device [07:42] OK, so it boots cleanly, and then can't find a device. [07:42] yes [07:42] * persia presses enter to get farther [07:42] Are you getting errors where it keeps sleeping for 5 seconds while checking device /dev/sd[abcd] for installation source? [07:42] yes [07:42] absolutely the same [07:43] OK. I've replicated that. [07:46] Hmm. It appears that the install sequence makes a number of assumptions about the HW present, and while it works for some devices, it doesn't for others. [07:47] ok [07:47] Essentially, it needs the booting USB key to have been mapped to /dev/sd[abcd] in order to proceed with the install, but that may not be the case for all HW environments. [07:47] so it wont work for USB???? [07:47] Well, at least it's not tuned for the board you have (not surprising, as this was tested only on the Crown Beach). [07:47] ohk [07:48] i have a crown beach board as well in office [07:48] but my concern is that in that as well there is no inbuilt storage [07:48] I'm going to again recommend booting off a USB image constructed from an i386 livecd (or booting off a USB optical drive), and seeing if you can determine where the devices are being mapped. [07:48] so will it install onto a USB [07:48] ok [07:49] I thought the crown beach did have a local storage interface (I don't have one, so I'm not sure). [07:49] all right [07:49] thanx a lot for ur help [07:50] You can definitely construct an image that installs onto USB, but you may have to modify Install.py to make sure that you are pointing at the right target. [07:50] Good luck with your solution. [07:50] install.py ?/? [07:50] where is that file??? [07:50] In the MIC source. [07:50] (moblin-image-creator) [07:50] yeah [07:51] and for MIC to work we need the 8.10 version of ubuntu rght cant get it off with just an old ubuntu rig? [07:54] I'd recommend getting the MIC from archive.mobile.ubuntu.com, as it has some improvements over that on archive.ubuntu.com for 8.04. [07:55] The 8.10 MIC is basically identical to the archive.mobile.ubuntu.com MIC, except it has been built for 8.10 instead of 8.04. [07:55] ok thnx a lot for ur tips [07:55] i'll try implementing them [07:56] once again thnx a lot [07:56] bala: I'm sorry I can't tell you more: I just don't have the HW to replicate that cleanly. [07:56] u have been helpful all this while answering patiently my qns that itself is enuf [15:25] persia: Hmmm you seem to claim that linux-lpiacompat works on i586+; I think lpia is i686, are you sure this is supported? [15:26] lool: amitk suggested it was, but I've never heard if it works or not from anyone with i586 HW. [15:27] And yes, lpia is i686, but apparently the linux-lpiacompat kernel was designed to create an lpia-like environment for non-lpia (or at least that was my understanding from what was said). [15:27] I'll see if I can find the log: I may be misinterpreting. [15:29] http://irclogs.ubuntu.com/2008/07/31/%23ubuntu-mobile.html [15:29] Ok; that was just in case; I don't strongly care, but was surprized by the discrepancy [15:30] persia: Sounds like you were intrepreting correctly to me; thanks for the inf [15:30] o [15:31] lool: No problem. It was a surprise to me as well. [15:32] * persia wants someone with an older Geode or a C7-M to report definitively "it works" or "it doesn't work". [15:33] I have some via here, but not C7 [15:35] Is it i586? [15:35] Yeah, at least [15:36] oh it's a c7 actually [15:36] When you have a chance, could you try booting the linux-lpiacompat kernel? If it works, I can feel confident with the advice I give Eee users. If it doesn't, I'll stop telling people it ought to work. [15:38] I'll need to unpack it and all some day; I'll do that ASAP but certainly not this WE [15:39] No huge rush. Be nice to know if it works, but generally those without lpia HW ought be testing in a VM. === Moot2 is now known as MootBot === Moot2 is now known as MootBot === Moot20 is now known as MootBot === Moot2 is now known as MootBot === birtz is now known as pyronik === emgent is now known as emgent` === emgent` is now known as emgent === Moot2 is now known as MootBot