Re: VCL xCAT2.x BladeCenter
hi Sunil, this file should exist in /sbin directory when rootimg.gz is unpacked . regards, prem. On Thu, Jun 30, 2011 at 6:48 PM, Sunil Venkatesh wrote: > Hi Josh/Prem, > > I was trying to get xcat working with the PPC. However, I couldn't find the > partimage-ng binary on the ramdisk. Is there any place I could find this > package? I have the partimage binaries that I downloaded off a website, but > couldn't find partimage-ng in that package. I am using statelite to capture > the image from the PPC blade. > > Regards, > Sunil > > On 6/9/11 4:58 PM, Prem Kumar wrote: >> >> Josh, >> >> got it to work!!! >> >> 'usepartimageng=0' set it to 1. >> >> never mind about the link to download partimage-ng, i was under the >> impression, that i will have to install partimage-ng, then looked >> around to realize it already exists in rootimg.gz. >> >> sincere thanks for patiently helping, now moving to next step! >> >> regards, >> prem. >> >> >> On Thu, Jun 9, 2011 at 3:50 PM, Prem Kumar >> wrote: >>> >>> hi Josh, >>> I need to add this to the documentation and should have mentioned it earlier. partimage can only capture a few file system types. I don't remember the exact list. I know it includes ext2/ext3 and ntfs. I also know that it cannot capture ext4 and anything involving LVM. That is part of why I included support for partimageng. If it doesn't natively know how to handle the file system, it will fall back to using dd to capture it. Search for 'usepartimageng=0' in the partimageng postscript and change it to =1 to switch to using partimageng. Is the file system to be captured either ext2 or ext3 and without having LVM invovled? >>> >>> file system involved is ext3 and without 'lvm'. so i am good on that. >>> >>> i could not find partimage-ng for download, could you point me to a >>> genuine source, sourceforege doesn't have it. >>> >>> following is the link to /tmp/partimageng.log http://pastebin.com/YrbmsF99 >>> >>> also noticed following error from partimage log in /var/log >>> >>> >> "[Can't read bit map block 0 from image]" >>> >>> looking out for that errors tells me that it could be because of >>> running partimage in 'gui' mode as opposed to batch, still looking for >>> more reasons. >>> >>> meanwhile i will look for other errors and wait for link to partimage-ng. >>> >>> regards, >>> prem. >>> >
Re: VCL xCAT2.x BladeCenter
Hi Josh/Prem, I was trying to get xcat working with the PPC. However, I couldn't find the partimage-ng binary on the ramdisk. Is there any place I could find this package? I have the partimage binaries that I downloaded off a website, but couldn't find partimage-ng in that package. I am using statelite to capture the image from the PPC blade. Regards, Sunil On 6/9/11 4:58 PM, Prem Kumar wrote: Josh, got it to work!!! 'usepartimageng=0' set it to 1. never mind about the link to download partimage-ng, i was under the impression, that i will have to install partimage-ng, then looked around to realize it already exists in rootimg.gz. sincere thanks for patiently helping, now moving to next step! regards, prem. On Thu, Jun 9, 2011 at 3:50 PM, Prem Kumar wrote: hi Josh, I need to add this to the documentation and should have mentioned it earlier. partimage can only capture a few file system types. I don't remember the exact list. I know it includes ext2/ext3 and ntfs. I also know that it cannot capture ext4 and anything involving LVM. That is part of why I included support for partimageng. If it doesn't natively know how to handle the file system, it will fall back to using dd to capture it. Search for 'usepartimageng=0' in the partimageng postscript and change it to =1 to switch to using partimageng. Is the file system to be captured either ext2 or ext3 and without having LVM invovled? file system involved is ext3 and without 'lvm'. so i am good on that. i could not find partimage-ng for download, could you point me to a genuine source, sourceforege doesn't have it. following is the link to /tmp/partimageng.log http://pastebin.com/YrbmsF99 also noticed following error from partimage log in /var/log >>"[Can't read bit map block 0 from image]" looking out for that errors tells me that it could be because of running partimage in 'gui' mode as opposed to batch, still looking for more reasons. meanwhile i will look for other errors and wait for link to partimage-ng. regards, prem.
Re: VCL xCAT2.x BladeCenter
hi Josh, i take back my statement below, saying that it would fail, if i don't have a symbolic link ext_sshd to sshd in the pam.d directory. * * my statement is only true if one follows the instructions * * at(http://www.kudos.be/multiple_sshd) to manually * * setup the external_sshd_config file.* * but wait why do it, when 'vcl' code already does it for you * * sorry for the confusion. regards, prem. On Fri, Jun 17, 2011 at 9:28 AM, Prem Kumar wrote: > hello Josh, > > indeed it was an sshd config issue: external_sshd_config > > also i would like to mention that, since i have 'PAM' enabled i had to > do the following as well to get it to work, and failed otherwise. > > cd /etc/pam.d > ln -s sshd ext_sshd > > i was scratching my head until i found the above solution at > http://www.kudos.be/multiple_sshd > > regards, > prem. > >> This sounds like an sshd config issue. I'd work with the >> external_sshd_config >> to find what is keeping you from being able to log in. Andy could confirm, >> but I'm pretty sure vcld sets that file up to allow password based logins. >> I'd compare external_sshd_config from the box that won't let you log in with >> sshd_config from a non-VCL machine that will let you use a password to log >> in. >> >> If you try sshing to it from another linux machine, you can use 'ssh -vvv' to >> get quite a bit of debugging info about the connection establishment. >> >> Josh >> >
Re: VCL xCAT2.x BladeCenter
hello Josh, indeed it was an sshd config issue: external_sshd_config also i would like to mention that, since i have 'PAM' enabled i had to do the following as well to get it to work, and failed otherwise. cd /etc/pam.d ln -s sshd ext_sshd i was scratching my head until i found the above solution at http://www.kudos.be/multiple_sshd regards, prem. > This sounds like an sshd config issue. I'd work with the external_sshd_config > to find what is keeping you from being able to log in. Andy could confirm, > but I'm pretty sure vcld sets that file up to allow password based logins. > I'd compare external_sshd_config from the box that won't let you log in with > sshd_config from a non-VCL machine that will let you use a password to log in. > > If you try sshing to it from another linux machine, you can use 'ssh -vvv' to > get quite a bit of debugging info about the connection establishment. > > Josh >
Re: VCL xCAT2.x BladeCenter
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Prem, This sounds like an sshd config issue. I'd work with the external_sshd_config to find what is keeping you from being able to log in. Andy could confirm, but I'm pretty sure vcld sets that file up to allow password based logins. I'd compare external_sshd_config from the box that won't let you log in with sshd_config from a non-VCL machine that will let you use a password to log in. If you try sshing to it from another linux machine, you can use 'ssh -vvv' to get quite a bit of debugging info about the connection establishment. Josh On Wednesday June 15, 2011, Prem Kumar wrote: > hello, > > one more issue still standing. > > with the swap of interfaces for 'IP0' and 'IP1'in 'Linux.pm' file i > was able login to the reserved node from the management node, but not > externally. > > with the current setup of image capture does it require that the end > user use public keys to login instead of password authentication? > > i am sure this is not a firewall issue, because when i use putty to > login, i get a prompt to type in the login username, and the moment i > type in username and hit enter, expecting to type in password, i get > the following error: > > "No supported authentication methods available" > > to me this seems to be a 'PAM' issue of 'sshd' config issue. > > any thoughts? or leads please? > > regards, > prem. > > On Tue, Jun 14, 2011 at 11:55 AM, Prem Kumar wrote: > > hi Andy, > > > > i will stick to this workaround for now, and this helps!!! > > > > if i were to run a post load or post reservation script, any idea > > where would that be? > > > > regards, > > prem. > > > > On Tue, Jun 14, 2011 at 10:00 AM, Andy Kurth wrote: > >> Hello Prem, > >> We're working to make the code more flexible regarding interface > >> names. For now, I think you're going to need to modify the code. > >> Open up lib/VCL/Module/OS/Linux.pm and find the generate_rc_local > >> subroutine. Find the following 2 lines and swap 'eth0' and 'eth1': > >> > >> push(@array2print, 'IP0=$(ifconfig eth0 | grep inet | awk \'{print > >> $2}\' | awk -F: \'{print $2}\')' . "\n"); > >> push(@array2print, 'IP1=$(ifconfig eth1 | grep inet | awk \'{print > >> $2}\' | awk -F: \'{print $2}\')' . "\n"); > >> > >> You will need to save a new revision of your image after making this > >> change. Once captured, the new revision should hopefully configure > >> SSH correctly. > >> > >> Hope this helps, > >> Andy > >> > >> On Mon, Jun 13, 2011 at 4:45 PM, Prem Kumar wrote: > >>> hello, > >>> > >>> every thing went well. have one minor correction to make. > >>> > >>> since my eth0 is public and eth1 is private, any idea on how could i > >>> guarantee that external_sshd_config listens to eth0 and gets the > >>> public ip and sshd_config listens to eth1 and gets private ip. > >>> > >>> because of the above switch of the ip's in sshd_config files, even > >>> though my reservation finishes it remains locked out. > >>> > >>> i also intend to provision using vmware, using the same management > >>> node and hoping that any changes made to get the above bare-metal > >>> install working will not affect the other. > >>> > >>> regards, > >>> prem. > >>> > >>> On Thu, Jun 9, 2011 at 4:58 PM, Prem Kumar wrote: > Josh, > > got it to work!!! > > 'usepartimageng=0' set it to 1. > > never mind about the link to download partimage-ng, i was under the > impression, that i will have to install partimage-ng, then looked > around to realize it already exists in rootimg.gz. > > sincere thanks for patiently helping, now moving to next step! > > regards, > prem. > > On Thu, Jun 9, 2011 at 3:50 PM, Prem Kumar wrote: > > hi Josh, > > > >> I need to add this to the documentation and should have mentioned it > >> earlier. partimage can only capture a few file system types. I > >> don't remember the exact list. I know it includes ext2/ext3 and > >> ntfs. I also know that it cannot capture ext4 and anything > >> involving LVM. That is part of why I included support for > >> partimageng. If it doesn't natively know how to handle the file > >> system, it will fall back to using dd to capture it. Search for > >> 'usepartimageng=0' in the partimageng postscript and change it to > >> =1 to switch to using partimageng. > >> > >> Is the file system to be captured either ext2 or ext3 and without > >> having LVM invovled? > > > > file system involved is ext3 and without 'lvm'. so i am good on that. > > > > i could not find partimage-ng for download, could you point me to a > > genuine source, sourceforege doesn't have it. > > > > following is the link to /tmp/partimageng.log > > > >> http://pastebin.com/YrbmsF99 > > > > also noticed following error from partimage log in /var/log > > > >
Re: VCL xCAT2.x BladeCenter
hello, one more issue still standing. with the swap of interfaces for 'IP0' and 'IP1'in 'Linux.pm' file i was able login to the reserved node from the management node, but not externally. with the current setup of image capture does it require that the end user use public keys to login instead of password authentication? i am sure this is not a firewall issue, because when i use putty to login, i get a prompt to type in the login username, and the moment i type in username and hit enter, expecting to type in password, i get the following error: "No supported authentication methods available" to me this seems to be a 'PAM' issue of 'sshd' config issue. any thoughts? or leads please? regards, prem. On Tue, Jun 14, 2011 at 11:55 AM, Prem Kumar wrote: > hi Andy, > > i will stick to this workaround for now, and this helps!!! > > if i were to run a post load or post reservation script, any idea > where would that be? > > regards, > prem. > > > On Tue, Jun 14, 2011 at 10:00 AM, Andy Kurth wrote: >> Hello Prem, >> We're working to make the code more flexible regarding interface >> names. For now, I think you're going to need to modify the code. >> Open up lib/VCL/Module/OS/Linux.pm and find the generate_rc_local >> subroutine. Find the following 2 lines and swap 'eth0' and 'eth1': >> >> push(@array2print, 'IP0=$(ifconfig eth0 | grep inet | awk \'{print >> $2}\' | awk -F: \'{print $2}\')' . "\n"); >> push(@array2print, 'IP1=$(ifconfig eth1 | grep inet | awk \'{print >> $2}\' | awk -F: \'{print $2}\')' . "\n"); >> >> You will need to save a new revision of your image after making this >> change. Once captured, the new revision should hopefully configure >> SSH correctly. >> >> Hope this helps, >> Andy >> >> >> On Mon, Jun 13, 2011 at 4:45 PM, Prem Kumar wrote: >>> hello, >>> >>> every thing went well. have one minor correction to make. >>> >>> since my eth0 is public and eth1 is private, any idea on how could i >>> guarantee that external_sshd_config listens to eth0 and gets the >>> public ip and sshd_config listens to eth1 and gets private ip. >>> >>> because of the above switch of the ip's in sshd_config files, even >>> though my reservation finishes it remains locked out. >>> >>> i also intend to provision using vmware, using the same management >>> node and hoping that any changes made to get the above bare-metal >>> install working will not affect the other. >>> >>> regards, >>> prem. >>> >>> On Thu, Jun 9, 2011 at 4:58 PM, Prem Kumar wrote: Josh, got it to work!!! 'usepartimageng=0' set it to 1. never mind about the link to download partimage-ng, i was under the impression, that i will have to install partimage-ng, then looked around to realize it already exists in rootimg.gz. sincere thanks for patiently helping, now moving to next step! regards, prem. On Thu, Jun 9, 2011 at 3:50 PM, Prem Kumar wrote: > hi Josh, > >> >> I need to add this to the documentation and should have mentioned it >> earlier. >> partimage can only capture a few file system types. I don't remember the >> exact list. I know it includes ext2/ext3 and ntfs. I also know that it >> cannot capture ext4 and anything involving LVM. That is part of why I >> included support for partimageng. If it doesn't natively know how to >> handle >> the file system, it will fall back to using dd to capture it. Search for >> 'usepartimageng=0' in the partimageng postscript and change it to =1 to >> switch >> to using partimageng. >> >> Is the file system to be captured either ext2 or ext3 and without having >> LVM >> invovled? > > file system involved is ext3 and without 'lvm'. so i am good on that. > > i could not find partimage-ng for download, could you point me to a > genuine source, sourceforege doesn't have it. > > following is the link to /tmp/partimageng.log >> http://pastebin.com/YrbmsF99 > > also noticed following error from partimage log in /var/log > > >> "[Can't read bit map block 0 from image]" > > looking out for that errors tells me that it could be because of > running partimage in 'gui' mode as opposed to batch, still looking for > more reasons. > > meanwhile i will look for other errors and wait for link to partimage-ng. > > regards, > prem. > >>> >> >
Re: VCL xCAT2.x BladeCenter
hi Andy, i will stick to this workaround for now, and this helps!!! if i were to run a post load or post reservation script, any idea where would that be? regards, prem. On Tue, Jun 14, 2011 at 10:00 AM, Andy Kurth wrote: > Hello Prem, > We're working to make the code more flexible regarding interface > names. For now, I think you're going to need to modify the code. > Open up lib/VCL/Module/OS/Linux.pm and find the generate_rc_local > subroutine. Find the following 2 lines and swap 'eth0' and 'eth1': > > push(@array2print, 'IP0=$(ifconfig eth0 | grep inet | awk \'{print > $2}\' | awk -F: \'{print $2}\')' . "\n"); > push(@array2print, 'IP1=$(ifconfig eth1 | grep inet | awk \'{print > $2}\' | awk -F: \'{print $2}\')' . "\n"); > > You will need to save a new revision of your image after making this > change. Once captured, the new revision should hopefully configure > SSH correctly. > > Hope this helps, > Andy > > > On Mon, Jun 13, 2011 at 4:45 PM, Prem Kumar wrote: >> hello, >> >> every thing went well. have one minor correction to make. >> >> since my eth0 is public and eth1 is private, any idea on how could i >> guarantee that external_sshd_config listens to eth0 and gets the >> public ip and sshd_config listens to eth1 and gets private ip. >> >> because of the above switch of the ip's in sshd_config files, even >> though my reservation finishes it remains locked out. >> >> i also intend to provision using vmware, using the same management >> node and hoping that any changes made to get the above bare-metal >> install working will not affect the other. >> >> regards, >> prem. >> >> On Thu, Jun 9, 2011 at 4:58 PM, Prem Kumar wrote: >>> Josh, >>> >>> got it to work!!! >>> >>> 'usepartimageng=0' set it to 1. >>> >>> never mind about the link to download partimage-ng, i was under the >>> impression, that i will have to install partimage-ng, then looked >>> around to realize it already exists in rootimg.gz. >>> >>> sincere thanks for patiently helping, now moving to next step! >>> >>> regards, >>> prem. >>> >>> >>> On Thu, Jun 9, 2011 at 3:50 PM, Prem Kumar wrote: hi Josh, > > I need to add this to the documentation and should have mentioned it > earlier. > partimage can only capture a few file system types. I don't remember the > exact list. I know it includes ext2/ext3 and ntfs. I also know that it > cannot capture ext4 and anything involving LVM. That is part of why I > included support for partimageng. If it doesn't natively know how to > handle > the file system, it will fall back to using dd to capture it. Search for > 'usepartimageng=0' in the partimageng postscript and change it to =1 to > switch > to using partimageng. > > Is the file system to be captured either ext2 or ext3 and without having > LVM > invovled? file system involved is ext3 and without 'lvm'. so i am good on that. i could not find partimage-ng for download, could you point me to a genuine source, sourceforege doesn't have it. following is the link to /tmp/partimageng.log > http://pastebin.com/YrbmsF99 also noticed following error from partimage log in /var/log >> "[Can't read bit map block 0 from image]" looking out for that errors tells me that it could be because of running partimage in 'gui' mode as opposed to batch, still looking for more reasons. meanwhile i will look for other errors and wait for link to partimage-ng. regards, prem. >>> >> >
Re: VCL xCAT2.x BladeCenter
Hello Prem, We're working to make the code more flexible regarding interface names. For now, I think you're going to need to modify the code. Open up lib/VCL/Module/OS/Linux.pm and find the generate_rc_local subroutine. Find the following 2 lines and swap 'eth0' and 'eth1': push(@array2print, 'IP0=$(ifconfig eth0 | grep inet | awk \'{print $2}\' | awk -F: \'{print $2}\')' . "\n"); push(@array2print, 'IP1=$(ifconfig eth1 | grep inet | awk \'{print $2}\' | awk -F: \'{print $2}\')' . "\n"); You will need to save a new revision of your image after making this change. Once captured, the new revision should hopefully configure SSH correctly. Hope this helps, Andy On Mon, Jun 13, 2011 at 4:45 PM, Prem Kumar wrote: > hello, > > every thing went well. have one minor correction to make. > > since my eth0 is public and eth1 is private, any idea on how could i > guarantee that external_sshd_config listens to eth0 and gets the > public ip and sshd_config listens to eth1 and gets private ip. > > because of the above switch of the ip's in sshd_config files, even > though my reservation finishes it remains locked out. > > i also intend to provision using vmware, using the same management > node and hoping that any changes made to get the above bare-metal > install working will not affect the other. > > regards, > prem. > > On Thu, Jun 9, 2011 at 4:58 PM, Prem Kumar wrote: >> Josh, >> >> got it to work!!! >> >> 'usepartimageng=0' set it to 1. >> >> never mind about the link to download partimage-ng, i was under the >> impression, that i will have to install partimage-ng, then looked >> around to realize it already exists in rootimg.gz. >> >> sincere thanks for patiently helping, now moving to next step! >> >> regards, >> prem. >> >> >> On Thu, Jun 9, 2011 at 3:50 PM, Prem Kumar wrote: >>> hi Josh, >>> I need to add this to the documentation and should have mentioned it earlier. partimage can only capture a few file system types. I don't remember the exact list. I know it includes ext2/ext3 and ntfs. I also know that it cannot capture ext4 and anything involving LVM. That is part of why I included support for partimageng. If it doesn't natively know how to handle the file system, it will fall back to using dd to capture it. Search for 'usepartimageng=0' in the partimageng postscript and change it to =1 to switch to using partimageng. Is the file system to be captured either ext2 or ext3 and without having LVM invovled? >>> >>> file system involved is ext3 and without 'lvm'. so i am good on that. >>> >>> i could not find partimage-ng for download, could you point me to a >>> genuine source, sourceforege doesn't have it. >>> >>> following is the link to /tmp/partimageng.log http://pastebin.com/YrbmsF99 >>> >>> also noticed following error from partimage log in /var/log >>> >>> >> "[Can't read bit map block 0 from image]" >>> >>> looking out for that errors tells me that it could be because of >>> running partimage in 'gui' mode as opposed to batch, still looking for >>> more reasons. >>> >>> meanwhile i will look for other errors and wait for link to partimage-ng. >>> >>> regards, >>> prem. >>> >> >
Re: VCL xCAT2.x BladeCenter
hello, every thing went well. have one minor correction to make. since my eth0 is public and eth1 is private, any idea on how could i guarantee that external_sshd_config listens to eth0 and gets the public ip and sshd_config listens to eth1 and gets private ip. because of the above switch of the ip's in sshd_config files, even though my reservation finishes it remains locked out. i also intend to provision using vmware, using the same management node and hoping that any changes made to get the above bare-metal install working will not affect the other. regards, prem. On Thu, Jun 9, 2011 at 4:58 PM, Prem Kumar wrote: > Josh, > > got it to work!!! > > 'usepartimageng=0' set it to 1. > > never mind about the link to download partimage-ng, i was under the > impression, that i will have to install partimage-ng, then looked > around to realize it already exists in rootimg.gz. > > sincere thanks for patiently helping, now moving to next step! > > regards, > prem. > > > On Thu, Jun 9, 2011 at 3:50 PM, Prem Kumar wrote: >> hi Josh, >> >>> >>> I need to add this to the documentation and should have mentioned it >>> earlier. >>> partimage can only capture a few file system types. I don't remember the >>> exact list. I know it includes ext2/ext3 and ntfs. I also know that it >>> cannot capture ext4 and anything involving LVM. That is part of why I >>> included support for partimageng. If it doesn't natively know how to handle >>> the file system, it will fall back to using dd to capture it. Search for >>> 'usepartimageng=0' in the partimageng postscript and change it to =1 to >>> switch >>> to using partimageng. >>> >>> Is the file system to be captured either ext2 or ext3 and without having LVM >>> invovled? >> >> file system involved is ext3 and without 'lvm'. so i am good on that. >> >> i could not find partimage-ng for download, could you point me to a >> genuine source, sourceforege doesn't have it. >> >> following is the link to /tmp/partimageng.log >>> http://pastebin.com/YrbmsF99 >> >> also noticed following error from partimage log in /var/log >> >> >> "[Can't read bit map block 0 from image]" >> >> looking out for that errors tells me that it could be because of >> running partimage in 'gui' mode as opposed to batch, still looking for >> more reasons. >> >> meanwhile i will look for other errors and wait for link to partimage-ng. >> >> regards, >> prem. >> >
Re: VCL xCAT2.x BladeCenter
Josh, got it to work!!! 'usepartimageng=0' set it to 1. never mind about the link to download partimage-ng, i was under the impression, that i will have to install partimage-ng, then looked around to realize it already exists in rootimg.gz. sincere thanks for patiently helping, now moving to next step! regards, prem. On Thu, Jun 9, 2011 at 3:50 PM, Prem Kumar wrote: > hi Josh, > >> >> I need to add this to the documentation and should have mentioned it earlier. >> partimage can only capture a few file system types. I don't remember the >> exact list. I know it includes ext2/ext3 and ntfs. I also know that it >> cannot capture ext4 and anything involving LVM. That is part of why I >> included support for partimageng. If it doesn't natively know how to handle >> the file system, it will fall back to using dd to capture it. Search for >> 'usepartimageng=0' in the partimageng postscript and change it to =1 to >> switch >> to using partimageng. >> >> Is the file system to be captured either ext2 or ext3 and without having LVM >> invovled? > > file system involved is ext3 and without 'lvm'. so i am good on that. > > i could not find partimage-ng for download, could you point me to a > genuine source, sourceforege doesn't have it. > > following is the link to /tmp/partimageng.log >> http://pastebin.com/YrbmsF99 > > also noticed following error from partimage log in /var/log > > >> "[Can't read bit map block 0 from image]" > > looking out for that errors tells me that it could be because of > running partimage in 'gui' mode as opposed to batch, still looking for > more reasons. > > meanwhile i will look for other errors and wait for link to partimage-ng. > > regards, > prem. >
Re: VCL xCAT2.x BladeCenter
hi Josh, > > I need to add this to the documentation and should have mentioned it earlier. > partimage can only capture a few file system types. I don't remember the > exact list. I know it includes ext2/ext3 and ntfs. I also know that it > cannot capture ext4 and anything involving LVM. That is part of why I > included support for partimageng. If it doesn't natively know how to handle > the file system, it will fall back to using dd to capture it. Search for > 'usepartimageng=0' in the partimageng postscript and change it to =1 to switch > to using partimageng. > > Is the file system to be captured either ext2 or ext3 and without having LVM > invovled? file system involved is ext3 and without 'lvm'. so i am good on that. i could not find partimage-ng for download, could you point me to a genuine source, sourceforege doesn't have it. following is the link to /tmp/partimageng.log > http://pastebin.com/YrbmsF99 also noticed following error from partimage log in /var/log >> "[Can't read bit map block 0 from image]" looking out for that errors tells me that it could be because of running partimage in 'gui' mode as opposed to batch, still looking for more reasons. meanwhile i will look for other errors and wait for link to partimage-ng. regards, prem.
Re: VCL xCAT2.x BladeCenter
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Prem, On Thursday June 09, 2011, Prem Kumar wrote: > hi Josh, > > significant progress ... although not fully successful. > > and very valuable tip from you!, this helped me quite a lot: "to aid > debugging enable remoteshell and removing the /sbin/reboot line at the > end" > > i want to make a note few things here that got me past a little > further in capture process, hence might be worth to see if i could > have done it better:: > > - since my primary nic was eth1, i greped-to-replace the extracted > files from initrd.img and rootimg.gz with eth1 as the primary nic. > > - make sure the node or the nodegroup being captured has appropriate > entries in the xcat table > postscripts.postscripts="syslog,remoteshell,syncfiles" or something > similar to suit individual needs. > > - it might be worth documenting the root password of the rootimg.gz > file system, unless one can replaced it with their own and re-create > the image, and then login using the root credentials when the capture > process fails at any point. worth looking is the /tmp/partimageng.log > of stateless image currently booted in. +++ not sure if this makes > sense, may be if the remoteshell is setup correctly we may not need > this? Our install ends up getting identity keys set up so that you don't have to use a password to log in. However, documenting the root password would be a good idea. > - i had to add the node being captured to my 'iptables' so that the > 'nfs' exports(managemen_node:/install) to write the image > at(/install/image/x86) is not blocked by the firewall on management > node. > > - in spite no_root_squash being enabled on management node nfs server, > 'root' account of the stateless image did not have the permissions to > write the image at /install/image/x86. it is believed that the version > of the 'nfs-client' in the stateless image, as i understand has some > issues. - workaround is to add the flag 'sec=sys' in the 'mount' > command in 'partimageng' script in folder /install/postscripts. > > - with the above adventure i got to a point where following > intermediate files were written to complete the capture process: > -rw-r--r-- 1 root root0 Jun 9 13:18 > vcl_rhel_template.img.capturefailed -rw--- 1 root root 849 Jun 9 > 13:18 vcl_rhel_template.gz.000 -rw-r--r-- 1 root root 259 Jun 9 > 13:24 vcl_rhel_template-sda.sfdisk -rw-r--r-- 1 root root 512 Jun 9 > 13:24 vcl_rhel_template-sda.mbr -rw--- 1 root root 10485760 Jun 9 > 13:24 pi43fec6f2.tmp I need to add this to the documentation and should have mentioned it earlier. partimage can only capture a few file system types. I don't remember the exact list. I know it includes ext2/ext3 and ntfs. I also know that it cannot capture ext4 and anything involving LVM. That is part of why I included support for partimageng. If it doesn't natively know how to handle the file system, it will fall back to using dd to capture it. Search for 'usepartimageng=0' in the partimageng postscript and change it to =1 to switch to using partimageng. Is the file system to be captured either ext2 or ext3 and without having LVM invovled? > - after few seconds, the '/sbin/partimage' saving the image, fails > with a status code 1. attached is the log file /tmp/partimageng.log > for additional information. partimage is definitely lacking in error messages. It does produce a log file somewhere under /var when it runs that might be useful. We have yet to figure out the constraints, but these ASF lists seem to eat many of the attachments people try to send. You might try just pasting relevant parts inline or posting it to pastebin.com and then including the link here. Josh - -- - --- Josh Thompson VCL Developer North Carolina State University my GPG/PGP key can be found at pgp.mit.edu -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.17 (GNU/Linux) iEYEARECAAYFAk3xHXAACgkQV/LQcNdtPQOQpgCdF7/paSYH4YkVuF9FBFpQmOWY NRUAn0ajFdczUmTeSLn4yiIigo6uPPOk =ZTGZ -END PGP SIGNATURE-
Re: VCL xCAT2.x BladeCenter
hi Josh, significant progress ... although not fully successful. and very valuable tip from you!, this helped me quite a lot: "to aid debugging enable remoteshell and removing the /sbin/reboot line at the end" i want to make a note few things here that got me past a little further in capture process, hence might be worth to see if i could have done it better:: - since my primary nic was eth1, i greped-to-replace the extracted files from initrd.img and rootimg.gz with eth1 as the primary nic. - make sure the node or the nodegroup being captured has appropriate entries in the xcat table postscripts.postscripts="syslog,remoteshell,syncfiles" or something similar to suit individual needs. - it might be worth documenting the root password of the rootimg.gz file system, unless one can replaced it with their own and re-create the image, and then login using the root credentials when the capture process fails at any point. worth looking is the /tmp/partimageng.log of stateless image currently booted in. +++ not sure if this makes sense, may be if the remoteshell is setup correctly we may not need this? - i had to add the node being captured to my 'iptables' so that the 'nfs' exports(managemen_node:/install) to write the image at(/install/image/x86) is not blocked by the firewall on management node. - in spite no_root_squash being enabled on management node nfs server, 'root' account of the stateless image did not have the permissions to write the image at /install/image/x86. it is believed that the version of the 'nfs-client' in the stateless image, as i understand has some issues. - workaround is to add the flag 'sec=sys' in the 'mount' command in 'partimageng' script in folder /install/postscripts. - with the above adventure i got to a point where following intermediate files were written to complete the capture process: -rw-r--r-- 1 root root0 Jun 9 13:18 vcl_rhel_template.img.capturefailed -rw--- 1 root root 849 Jun 9 13:18 vcl_rhel_template.gz.000 -rw-r--r-- 1 root root 259 Jun 9 13:24 vcl_rhel_template-sda.sfdisk -rw-r--r-- 1 root root 512 Jun 9 13:24 vcl_rhel_template-sda.mbr -rw--- 1 root root 10485760 Jun 9 13:24 pi43fec6f2.tmp - after few seconds, the '/sbin/partimage' saving the image, fails with a status code 1. attached is the log file /tmp/partimageng.log for additional information. need more of your assistance? regards, prem.
Re: VCL xCAT2.x BladeCenter
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Prem, Good to hear you are making progress. On Wednesday June 08, 2011, Prem Kumar wrote: > hi Josh, > > made some progress with 'os' and 'arch' in nodetype set to 'image' and > 'x86', yet i am missing something. > > kindly could you confirm where i went wrong from below in reference to > xcat and partimage setup: > > - successfully installed a compute node with pxe boot and 'ks' > template located under: /install/custom/install/rh > > - copied vmlinuz and initrd.img from > here(https://cwiki.apache.org/VCLDOCS/adding-support-for-partimage-and-part > image-ng-to-xcat-2x-unofficial.html) to /tftpboot/xcat/image/x86/ > > - also copied rootimg.gz to /install/image/x86/installer_files/ > > ++since the above images were built with eth0 as the primary nic, > i changed it to eth1 because that being my primary nic. I'm glad you found that. I had forgotten they have hard coded NICs. I'll look in to getting that info from xcat. > ++ not sure if the rootimg.gz here has to be the cpio image from > the install of the compute node in the first step? and the vmlinuz and > initrd.img from the disk or rhel used to install the compute node. I'm not clear what you are asking here. You should be able to use the rootimg.gz from the "unofficial" wiki page you listed above. > - then set 'os' and 'arch' in nodetype table to 'image' and 'x86' > respectively > > - next typed in the following command: > > nodeset image > ++ at this point it complained that it could not find: > /opt/xcat/share/xcat/install/image/createimage.x86.tmpl , i was hoping > my file /opt/xcat/share/xcat/install/image/vcl_rhel_template.tmpl with > this single line(#INCLUDE:../scripts/post.rh#) in it should have done > the job. not sure why it only needs createimage.x86.tmpl ? It looks like I left this out of the docs. For pushing images, you need one of these under /opt/xcat/share/xcat/install/image (somewhat documented): $profile.$arch.tmpl $profile.tmpl default.tmpl For capturing images, you need this under /opt/xcat/share/xcat/install/image (undocumented): createimage.$arch.tmpl That file should only have the following single line: #INCLUDE:../scripts/post.rh# It would make sense to update the partimageng.pm module to have capturing requirements similar to pushing requirements and allow $profile.$arch.tmpl and $profile.tmpl to work. I'll try to get that updated in the next few weeks. > after creating the required(createimage.x86.tmpl) file, and the typing > the above command creates appropriate pxe configuration files in > /tftpboot/pxelinux.cfg/ with the url and action to save the image etc, > which seems right. > > - now type in the following command to boot the node so as to save the > image upon reboot > rpower boot > > at this point if everything is right, following should happen > > > the stateless image will be used to boot in initial ram disk > > > > then use the postinstall partimage script to save the image of the > > current installation of the 'os' onto the location on management node. > > +++ here i don't see when does one gets the post install > scripts(partimage) from management node? Looking in rootimg.gz, you'll find /etc/rc3.d/S84xcatpostinit which is a symlink to /etc/init.d/xcatpostinit which runs /opt/xcat/xcatdsklspost which uses wget to download all of /install/postscripts from the management node. Then it executes the getpostscript.awk script to generate a primary postscript file. This file should contain various things, one of which is the partimageng script. It runs this primary postscript, thus running the partimageng script. To aid in debugging, if you configure the node to also have the remoteshell postscript run, you'll be able to ssh in to it while the stateless image is booted. There may be some things you need to set up to get the remoteshell postscript working. Also, if the node reboots too quickly for you to poke around, you can comment out the /sbin/reboot line at the end of the partimageng postscript. > > this should finish the capture process > > hope this is correct? > > regards, > prem. Josh - -- - --- Josh Thompson VCL Developer North Carolina State University my GPG/PGP key can be found at pgp.mit.edu -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.17 (GNU/Linux) iEYEARECAAYFAk3wvowACgkQV/LQcNdtPQMItACfQ3yl7tA6oyoWVtTX7de6BKYT SFAAmwWV/rZV4ZakDLIqV7XFkQhgTdi6 =hzwX -END PGP SIGNATURE-
Re: VCL xCAT2.x BladeCenter
hi Josh, made some progress with 'os' and 'arch' in nodetype set to 'image' and 'x86', yet i am missing something. kindly could you confirm where i went wrong from below in reference to xcat and partimage setup: - successfully installed a compute node with pxe boot and 'ks' template located under: /install/custom/install/rh - copied vmlinuz and initrd.img from here(https://cwiki.apache.org/VCLDOCS/adding-support-for-partimage-and-partimage-ng-to-xcat-2x-unofficial.html) to /tftpboot/xcat/image/x86/ - also copied rootimg.gz to /install/image/x86/installer_files/ ++since the above images were built with eth0 as the primary nic, i changed it to eth1 because that being my primary nic. ++ not sure if the rootimg.gz here has to be the cpio image from the install of the compute node in the first step? and the vmlinuz and initrd.img from the disk or rhel used to install the compute node. - then set 'os' and 'arch' in nodetype table to 'image' and 'x86' respectively - next typed in the following command: nodeset image ++ at this point it complained that it could not find: /opt/xcat/share/xcat/install/image/createimage.x86.tmpl , i was hoping my file /opt/xcat/share/xcat/install/image/vcl_rhel_template.tmpl with this single line(#INCLUDE:../scripts/post.rh#) in it should have done the job. not sure why it only needs createimage.x86.tmpl ? after creating the required(createimage.x86.tmpl) file, and the typing the above command creates appropriate pxe configuration files in /tftpboot/pxelinux.cfg/ with the url and action to save the image etc, which seems right. - now type in the following command to boot the node so as to save the image upon reboot rpower boot at this point if everything is right, following should happen > the stateless image will be used to boot in initial ram disk > then use the postinstall partimage script to save the image of the current > installation of the 'os' onto the location on management node. +++ here i don't see when does one gets the post install scripts(partimage) from management node? > this should finish the capture process hope this is correct? regards, prem.
Re: VCL xCAT2.x BladeCenter
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Wednesday June 08, 2011, Prem Kumar wrote: > hi Josh, kindly bare with other inline questions before i try out your > suggestions - > > > > 'vcl_rhel_node_template' then i need to have the following file > > > under $installroot/image/$arch/ > > > > > > vcl_rhel_node_template.img > > > vcl_rhel_node_template.gz > > > vcl_rhel_node_template-parta1.gz > > > > The above code means that you must have at least one of those files > > present in $installroot/image/$arch/. Those are the disk image files > > created by partimageng (any .img files) or partimage (any .gz files). > > The partimage stuff documented at the "unofficial" page always uses x86 > > as for $arch. It will still handle imaging/restoring x86_64 images. > > Since both x86 and x86_64 were handled the same way, I never bothered > > making it distinguish between them. So, all images captured with > > partimage using those instruction will be under $installroot/image/x86/ > > as i understand: > essentially files vcl_rhel_node_template.img 'or' > vcl_rhel_node_template.gz gets created for the first time after i have > successfully captured an image using the partimageng or partimage > respectfully? Correct. > and this is done using the command 'nodeset image' Correct. > also do you think having ssh communicate to private management node > over eth1, and compute nodes accessible to end users with ext_ssh > over eth0, an issue with the 'vcl code' at any point? I don't think it will be a problem. At one point in the past, eth0 had to be used for the private network and eth1 for the public. However, I believe all places that were hard coded to one or the other have been updated to be configurable. Josh - -- - --- Josh Thompson VCL Developer North Carolina State University my GPG/PGP key can be found at pgp.mit.edu -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.17 (GNU/Linux) iEYEARECAAYFAk3vmIcACgkQV/LQcNdtPQP9MQCfUbNU+p9i5994kzm7PH47x1av 5T8AmgOAfuUgtS6evSG2IBN7pBOSaQsC =1qDh -END PGP SIGNATURE-
Re: VCL xCAT2.x BladeCenter
hi Josh, kindly bare with other inline questions before i try out your suggestions - > > 'vcl_rhel_node_template' then i need to have the following file > > under $installroot/image/$arch/ > > > > vcl_rhel_node_template.img > > vcl_rhel_node_template.gz > > vcl_rhel_node_template-parta1.gz > > The above code means that you must have at least one of those files present in > $installroot/image/$arch/. Those are the disk image files created by > partimageng (any .img files) or partimage (any .gz files). The partimage > stuff documented at the "unofficial" page always uses x86 as for $arch. It > will still handle imaging/restoring x86_64 images. Since both x86 and x86_64 > were handled the same way, I never bothered making it distinguish between > them. So, all images captured with partimage using those instruction will be > under $installroot/image/x86/ > as i understand: essentially files vcl_rhel_node_template.img 'or' vcl_rhel_node_template.gz gets created for the first time after i have successfully captured an image using the partimageng or partimage respectfully? and this is done using the command 'nodeset image' also do you think having ssh communicate to private management node over eth1, and compute nodes accessible to end users with ext_ssh over eth0, an issue with the 'vcl code' at any point? regards, prem.
Re: VCL xCAT2.x BladeCenter
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Prem, My responses are inline. On Wednesday June 08, 2011, Prem Kumar wrote: > hi Josh, > > no i do not have the 'os' filed set to 'image' in the nodetype table. > also i do not have the 'arch' field set to 'x86', instead i have it set to > 'x86_64'. > > this probably explains why it is not progressing. Okay. Let us know how things progress when you set the values as described. > just so i understand this correctly, this has been my approach, can you > confirm if i am following the right trial of path. > > - first i got my xcat to work, so that i can install an os on baremetal > using pxe boot with kickstart templates. > - second i got the vcl code setup with unoffical-partimage-setup > information. > - third i followed the instructions at > https://cwiki.apache.org/VCL/create-a-linux-base-image.html to capture the > baremetal install of the 'os' on a blade. That is the correct approach. However, I would not try using 'vcld -setup' until you can successfully capture an image using 'nodeset image'. > back to the partimage plugin: > > i read around line number 142: > unless (-r "$installroot/image/$arch/$profile.img" or > -r "$installroot/image/$arch/$profile.gz" or > -r "$installroot/image/$arch/$profile-parta1.gz") > > does the above imply that if my 'profile' in nodetype table is set to > 'vcl_rhel_node_template' then i need to have the following file > under $installroot/image/$arch/ > > vcl_rhel_node_template.img > vcl_rhel_node_template.gz > vcl_rhel_node_template-parta1.gz The above code means that you must have at least one of those files present in $installroot/image/$arch/. Those are the disk image files created by partimageng (any .img files) or partimage (any .gz files). The partimage stuff documented at the "unofficial" page always uses x86 as for $arch. It will still handle imaging/restoring x86_64 images. Since both x86 and x86_64 were handled the same way, I never bothered making it distinguish between them. So, all images captured with partimage using those instruction will be under $installroot/image/x86/ As a side note, there is another thread where someone is working on making partimage work for ppc. For that, it probably would be good to have it distinguish between x86 and ppc. Josh > ?? > > regards, > prem > > > Prem, > > > > Make sure you have partimageng.pm in /opt/xcat/lib/perl/xCAT_plugin. > > > > The "os" field in nodetype needs to be "image" and the "arch" field needs > > to > > be "x86". > > > > Can you confirm that is what you have specified in nodetype? > > > > Josh > > > > On Tuesday June 07, 2011, Prem Kumar wrote: > > > hello, > > > > > > i am trying to get bare metal installs to work with vcl2.2 and xcat 2.x > > > > > using the info provided at: > > https://cwiki.apache.org/VCL/adding-support-for-partimage-and-partimage-n > > g- > > > > > to-xcat-2x-unofficial.html > > > > > > my xcat installation works fine: i have successfully installed rhel > > > nodes using the following commands > > > nodeset install=Template > > > rpower boot > > > > > > > > > now when i want to first image and then install using the partimage i > > > am having issues as follows: > > > > > > i am running into following error the moment I type in: nodeset > > > image even after rebooting the management node there has been no change > > > > in > > > > > this error. > > > > > > > > > baremetal-chassis1-13: Unable to identify plugin for this command, > > > check relevant tables: nodetype.os > > > Error: Some nodes failed to set up image resources, aborting > > > > > > > > > to my understanding, for some reason partimage plugin is not being fond > > > > or > > > > > recognized by xcat. > > > > > > can anybody please help. > > > > > > regards, > > > prem. > > > > - -- > > - --- > > Josh Thompson > > VCL Developer > > North Carolina State University - -- - --- Josh Thompson VCL Developer North Carolina State University my GPG/PGP key can be found at pgp.mit.edu -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.17 (GNU/Linux) iEYEARECAAYFAk3vgokACgkQV/LQcNdtPQP7bQCfdxoh70lvoNtA1PxnL2T86P6h LOoAn2LcYnMe0CsnkzzG5HGSI+ze0OnK =ad1X -END PGP SIGNATURE-
Re: VCL xCAT2.x BladeCenter
hi Josh, no i do not have the 'os' filed set to 'image' in the nodetype table. also i do not have the 'arch' field set to 'x86', instead i have it set to 'x86_64'. this probably explains why it is not progressing. just so i understand this correctly, this has been my approach, can you confirm if i am following the right trial of path. - first i got my xcat to work, so that i can install an os on baremetal using pxe boot with kickstart templates. - second i got the vcl code setup with unoffical-partimage-setup information. - third i followed the instructions at https://cwiki.apache.org/VCL/create-a-linux-base-image.html to capture the baremetal install of the 'os' on a blade. back to the partimage plugin: i read around line number 142: unless (-r "$installroot/image/$arch/$profile.img" or -r "$installroot/image/$arch/$profile.gz" or -r "$installroot/image/$arch/$profile-parta1.gz") does the above imply that if my 'profile' in nodetype table is set to 'vcl_rhel_node_template' then i need to have the following file under $installroot/image/$arch/ vcl_rhel_node_template.img vcl_rhel_node_template.gz vcl_rhel_node_template-parta1.gz ?? regards, prem -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Prem, > > Make sure you have partimageng.pm in /opt/xcat/lib/perl/xCAT_plugin. > > The "os" field in nodetype needs to be "image" and the "arch" field needs > to > be "x86". > > Can you confirm that is what you have specified in nodetype? > > Josh > > On Tuesday June 07, 2011, Prem Kumar wrote: > > hello, > > > > i am trying to get bare metal installs to work with vcl2.2 and xcat 2.x > > using the info provided at: > > > https://cwiki.apache.org/VCL/adding-support-for-partimage-and-partimage-ng- > > to-xcat-2x-unofficial.html > > > > my xcat installation works fine: i have successfully installed rhel nodes > > using the following commands > > nodeset install=Template > > rpower boot > > > > > > now when i want to first image and then install using the partimage i am > > having issues as follows: > > > > i am running into following error the moment I type in: nodeset > > image even after rebooting the management node there has been no change > in > > this error. > > > > > > baremetal-chassis1-13: Unable to identify plugin for this command, check > > relevant tables: nodetype.os > > Error: Some nodes failed to set up image resources, aborting > > > > > > to my understanding, for some reason partimage plugin is not being fond > or > > recognized by xcat. > > > > can anybody please help. > > > > regards, > > prem. > - -- > - --- > Josh Thompson > VCL Developer > North Carolina State University > >
Re: VCL xCAT2.x BladeCenter
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Prem, Make sure you have partimageng.pm in /opt/xcat/lib/perl/xCAT_plugin. The "os" field in nodetype needs to be "image" and the "arch" field needs to be "x86". Can you confirm that is what you have specified in nodetype? Josh On Tuesday June 07, 2011, Prem Kumar wrote: > hello, > > i am trying to get bare metal installs to work with vcl2.2 and xcat 2.x > using the info provided at: > https://cwiki.apache.org/VCL/adding-support-for-partimage-and-partimage-ng- > to-xcat-2x-unofficial.html > > my xcat installation works fine: i have successfully installed rhel nodes > using the following commands > nodeset install=Template > rpower boot > > > now when i want to first image and then install using the partimage i am > having issues as follows: > > i am running into following error the moment I type in: nodeset > image even after rebooting the management node there has been no change in > this error. > > > baremetal-chassis1-13: Unable to identify plugin for this command, check > relevant tables: nodetype.os > Error: Some nodes failed to set up image resources, aborting > > > to my understanding, for some reason partimage plugin is not being fond or > recognized by xcat. > > can anybody please help. > > regards, > prem. - -- - --- Josh Thompson VCL Developer North Carolina State University my GPG/PGP key can be found at pgp.mit.edu -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.17 (GNU/Linux) iEYEARECAAYFAk3ucn4ACgkQV/LQcNdtPQPmCgCfYrJfhJKSYIOrX5+U9Kepi3qn CqEAn37PQlTGO6SE22g4lLXetJW1Mpfs =PaiS -END PGP SIGNATURE-
VCL xCAT2.x BladeCenter
hello, i am trying to get bare metal installs to work with vcl2.2 and xcat 2.x using the info provided at: https://cwiki.apache.org/VCL/adding-support-for-partimage-and-partimage-ng-to-xcat-2x-unofficial.html my xcat installation works fine: i have successfully installed rhel nodes using the following commands nodeset install=Template rpower boot now when i want to first image and then install using the partimage i am having issues as follows: i am running into following error the moment I type in: nodeset image even after rebooting the management node there has been no change in this error. > baremetal-chassis1-13: Unable to identify plugin for this command, check relevant tables: nodetype.os Error: Some nodes failed to set up image resources, aborting to my understanding, for some reason partimage plugin is not being fond or recognized by xcat. can anybody please help. regards, prem.