[TriLUG] e2fsck under cron gets retcode=8 operational error

[Ken Mink]

On 9/19/12 9:16 AM, Joseph Mack NA3T wrote:
> On Wed, 19 Sep 2012, Thomas Gardner wrote:
>
>> [...whatever...] You don't always have to break things to make them 
>> more usable for more use cases.  Never mind. I'll just put it on my 
>> long list of things I want to do ``some day''.  If you could come up 
>> with a list of pointers to those other forums where you saw specific 
>> cases where this was messing people up, I can use that as supporting 
>> evidence.
>
> I'll be more than impressed if you can get this fixed. It would show 
> me that there's more hope of getting the world to work than ever I 
> thought.
>
> I have a map generator that makes maps for ham radio operators (code 
> first released in 1994). Some of the data comes from ham sites (eg 
> lists of beacons, logs from contests). Almost none of it is machine 
> readable. Comments and notes are spread throughout, like it was a 
> letter to a friend. I contacted the list maintainers, asking them to 
> use a machine readable format. I suggested a few, but was quite happy 
> to use a format of their choosing. I wanted my map generator to be 
> able to download new versions of the files automatically and map them 
> without human intervention. I didn't want my map generator crashing 
> because it was trying to plot a corrupt file. Most of the list 
> maintainers didn't reply. The most encouraging reply was that the list 
> maintainer thought it a good idea, but he wasn't going to be the first.
>
> I tried to plot the activity in contests, where you look for unusual 
> propagation that will only be seen when large numbers of hams are 
> operating at the same time. The contest organisers won't let you have 
> the data submitted for the contest. The contest is not auditable. 
> Neither can you learn from it. The ARRL's policy is that the contest 
> winners have put a lot of work into getting to where they are, and to 
> have people learn what they know, would negate their efforts.
>
> I decided if these people really didn't want their data plotted on my 
> map generator, then I wouldn't plot it.
>
> The last ham radio event I plotted was the biggest aurora opening in 
> the last sunspot cycle, in 2000. After that I gave up on ham radio and 
> put my efforts into Linux.
>
> In the process of collecting data to be plotted on my map generator, I 
> found that hams were computer illiterate. They're just a bunch of 
> windows or Mac users. I would ask for the files in ascii format and 
> get binary doc files from Macs etc. One person sent me jpg photo of 
> his log page. I talked to one of these people face-to-face and he was 
> adamant that the file was ascii; he saw characters on his screen and 
> not binary and that was all there was to it. When I started ham radio 
> (50's and 60's), the ethic was build it yourself and understand 
> everything you're doing. When Linux arrived in the '90's I recognised 
> it as another ham radio. However hams haven't moved. AFAICT they are 
> still back in the '50s and '60s. Many of them just want to operate 
> their rigs. This is fine, people can do this if they want. But when I 
> go to ham radio conferences, where people are doing the cutting edge 
> ham radio (eg moonbounce), they are still using windows for their 
> computing. They don't see a connection between FOSS and ham radio.
>
> As a side note, people living in the US should be very glad of some of 
> the US Govt's policies. Data the US Govt collects (eg location of post 
> offices, county borders, TV and radio station frequencies etc) using 
> taxes from the citizens, is by virtue of the FoI act, freely 
> available. I use it on my map generator. Other countries (even 
> civilised places like Canada) don't make it available at all, or 
> charge you large amounts of money for it and place restrictions on 
> what you can do with it. It's one very nice thing about the US Govt.
>
> In the fsck case, I googled for
>
> e2fsck operational error
>
> where "operational error" is the text version of the error code.
>
> I came up with
>
> http://www.techtalkz.com/suse-linux/183070-problems-fsck-status-0x8-boot.html 
>
> http://www-01.ibm.com/support/docview.wss?uid=swg21387564
> http://www.linuxquestions.org/questions/ubuntu-63/fsck-died-with-exit-status-8-a-499851/ 
>
> http://forums.gentoo.org/viewtopic-t-918740-start-0.html
>
> Joe
>
It took me a while to get caught up on this thread. I didn't see it 
mentioned before, but are you using the '-y' switch? When I read the 
original post it just sounded like fsck was asking a question and was 
getting clobbered by cron because it sat too long.

Ken