<div dir="ltr"><br><div class="gmail_extra"><div class="gmail_quote">On Fri, Apr 12, 2013 at 2:17 PM, Rich West <span dir="ltr"><<a href="mailto:Rich.West@wesmo.com" target="_blank">Rich.West@wesmo.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div>On 04/12/2013 01:13 PM, Mike Perkins wrote:<br>
> On 12/04/13 17:45, Rich West wrote:<br>
>> This past weekend I finally bit the bullet and moved from 0.25 to 0.26.<br>
>> As simple as that sounds, it wasn't. :( I had to upgrade from Fedora 16<br>
>> to Fedora 18 on all of my frontends as well as my backend since I try to<br>
>> keep up with the RPM distribution, and, in the process, I was replacing<br>
>> my frontend hardware (machine and universal remotes).<br>
>><br>
>> Anyhow, the OS upgrade went pretty well (that was a shock), and, after a<br>
>> few hiccups with the mythbackend, things have stabilized out and the WAF<br>
>> has been steadily coming back.<br>
>><br>
>> Since the upgrade, I have been cleaning up missing recordings and stuff<br>
>> out of the database (had a drive failure back in January which resulted<br>
>> in a lot of lost recordings) to make sure it is clean. Yesterday, I was<br>
>> positive I had cleaned out all of the remaining stuff.<br>
>><br>
>> Since I monitor this system via Cacti for system stats as well as mythtv<br>
>> stats, I noticed a steady increase in the number of errors it is<br>
>> reporting despite all of the old entries pointing to non-existant files<br>
>> being purged. In looking at the backend logs, I'm seeing:<br>
>><br>
>> 2013-04-12 10:00:27.377894 E [18735/21090] Commflag_118942<br>
>> programinfo.cpp:2284 (GetPlaybackURL) -<br>
>> ProgramInfo(1292_20130412140000.mpg): GetPlaybackURL:<br>
>> '1292_20130412140000.mpg' should be local, but it can not be found.<br>
>><br>
>> Since 7am yesterday, there are 456 unique instances of those errors in<br>
>> my backend log. And those recordings are current (judging by the<br>
>> filename ####_YYYYMMDD######.mpg).. so why would they disappear? I<br>
>> checked on my two volumes (I have two separate 500GB disks in a storage<br>
>> group) and those files don't exist.<br>
>><br>
>> I'm not sure where else to look for narrowing down this particular<br>
>> problem. Thoughts?<br>
>><br>
>> -Rich<br>
>><br>
>> Some stats, if they are helpful to anyone:<br>
>> Mythtv 0.26<br>
>> 3 PVR150's via S-video to DirectTV SD boxes with USB connections to<br>
>> change channels<br>
>> 1 dual-tuner HD Homerun<br>
>> Dual-core AMD64 X2 4000+<br>
>> 2GB ram<br>
>> 2 500GB drives in default storage group<br>
>> 2 200GB drives in a mirrored root pair<br>
>> mysql database is local<br>
>> CPU utilization is low (benefit of the tuners)<br>
>><br>
> I suspect this is to do with the change to UTC. Try looking for<br>
> filenames on your disk which are however many hours away your timezone<br>
> is from the filename.<br>
><br>
> I thought the database upgrade from 0.25 to 0.26 would have fixed<br>
> this, but perhaps there are corner cases relating to your previous<br>
> problems.<br>
><br>
<br>
</div></div>Being in EST (EDT at the moment, so it's UTC/GMT-4), I did hunt for<br>
recent files, but unfortunately came up empty. "ls -latR /volumename |<br>
head" for each volume didn't show anything created within the last three<br>
hours. :( For example, with the file above, there hasn't been any file<br>
created today starting with 1292_20130412.<br>
<br>
Hmm.. wait a sec.. I bet it is permissions related. Using ATrpms,<br>
mythbackend ran as root. Using RPMfusion, it's running at mythtv.<br>
Despite the video devices being world read/write and the storage group<br>
also being world read/write, it's still a problem. Restarting it as<br>
root seemed to start recordings again! Now I get to attack any other<br>
issues that come up. :)<br>
<span><font color="#888888"><br>
-Rich<br>
</font></span><div><div>_______________________________________________<br>
mythtv-users mailing list<br>
<a href="mailto:mythtv-users@mythtv.org" target="_blank">mythtv-users@mythtv.org</a><br>
<a href="http://www.mythtv.org/mailman/listinfo/mythtv-users" target="_blank">http://www.mythtv.org/mailman/listinfo/mythtv-users</a><br>
</div></div></blockquote></div><br><br></div><div class="gmail_extra">Rich:<br><br></div><div class="gmail_extra">I am staring down the same barrel -- an old Fedora 16 system on 0.25.1 that needs to move to F18 and 0.26. Having previously had unsuccessful attempts to use preupgrade to move from F16 to F17 (and, I understand that you can't move directly from F16 to F18 using either preupgrade or fedup), what did you do? Can you give me a brief outline of your process? <br>
<br>My gut says that the best course is to simply do a clean install of F18, but having used the new anaconda (and its horrible disk partitioner), I'm worried that I'll mess up my partitions (I have a separate /home and multiple /video# partitions). I'm tempted to do the clean install of F17 and then use fedup to move to F18 -- or perhaps wait for F19, since F18 isn't anything special.<br>
<br></div><div class="gmail_extra">As for myth, my assumption is the same as always -- backup the database, do the upgrade and use the backup if there are problems. My database has been successfully upgraded each time since 0.12, so I'm not overly concerned about that part.<br>
</div><div class="gmail_extra"><br>Any thoughts?<br>
<br>Thanks,<br><br>George<br></div></div>