Best posts made by Junkhacker

something i’ve done is to just leave windows running on a vm, getting windows updates automatically, and having a scheduled upload task that runs every week. you’ll want to not install the vm tools for the machine, since they can seem to cause some odd behavior when the image gets deployed back to real hardware

it does work with an external keyboard…
is that a driver related issue?

almost certainly, yes.
personally, i’d just use an external keyboard instead of troubleshooting further…

@Junkhacker interesting. no problems with images that are told not to generate checksums. no problem with images with checksums if i remove the --ignore_crc parameter.

@Wayne-Workman

Yes. But the version that JunkHacker referenced is probably already outdated. It’s TWO hours old!

actually, it’s about 6 days old, but it’s the version i’m running in production and I’ve verified that the functions of fog i consider critical (capture/deploy windows 7-10, snapins, hostname changer/domain join, etc) are working properly (for me, at least). I run separate “dev” and “production” servers and only move builds into production that I’ve verified are relatively stable, and i don’t like recommending builds i haven’t tested.
It’s great when people want to install the latest dev builds and test and use them, and we couldn’t make much progress without that and for the most part it works fine. but in my opinion, i think should recommend builds to people that we know are relatively stable. especially when people are still learning how fog works.

@cmurray139 i’m not sure. I don’t have much experience with devices that don’t have Ethernet built in. if you search the forums, you might find someone who’s already solved the problem though.

@Wayne-Workman said in Odd pending host issues:

@x23piracy Yes, update again. Patches are pushed many, many times a day. The senior developer has already said in this thread that he thinks he’s fixed the issue. So to get the patch, you must update.

actually, Tom’s comment and the original problem listed in the thread are probably unrelated since x23piracy is resurrecting a year old thread…

@zfeng it sounds like George’s method of creating a file in /etc/init.d on the target disk would work well for you. you can find info on creating postdownload scripts by searching on the forums

it looks to me like maybe your ftp root isn’t right

@zfeng you could do something like wget -o /root/script.sh <address>/script.sh; echo “Computer=$hostname /root/script.sh” > <drive mount point>/etc/init.d/oneshot.sh

something like that.

you won’t lose your images on an upgrade, and what pxe problems did you have? i’m sure we can fix them

@zfeng in a postdownload script, so that it will run when the imaging process completes but before the imaging task completes

@prestop said in Hosts not imaging - HELP:

@prestop
Something else changed as well:
the one image that had a size listed for client disk now also says “0.00iB”

… you haven’t been mixing up the upload/download have you?
upload to server, download to host (this naming convention has been changed in trunk)
i’m just worried that you’ve uploaded blank images over your existing ones

@Wayne-Workman i realize i’m coming into this really late, but why not randomize what day of the week we collect the data as well, and we can just look at the window of “last 7 days” when we look at the data?

since you now have a backup of the server you can recover from, i can’t see any reason not to try the dev build (commonly referred to on the forums as “trunk”)

@Wayne-Workman how about “install survey”

@Quazz said in Advanced Menu Issue(s):

This is most likely caused by an invalid formatted advanced menu.

that is why i wanted to see the output of that url

@Tom-Elliott i think we should default to Y/n instead of y/N

most people don’t care one way or the other and will just click through, let the people who care chose not to send it

@prestop oh yeah, btw, the trunk version of fog is faster. depending on circumstances, it can be a LOT faster. lol

on the subject of “not using nfsv3” i have an idea on how to reimplement torrent-casting. clients receiving images wouldn’t need file level access at all (though uploads would still need nfs).

would there be any interest?

@fry_p i wouldn’t worry. frankly, i don’t think the graph is the most accurate thing in the world due to sample rate. it’s mostly useful to be able to tell if you’re saturating your connection to the server