[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: Fedora downtime

Fulko Hew sita aero wrote:

fedora-test-list-bounces redhat com wrote on 10/25/2006 10:34:24 AM:

Could anybody put here at least small note what's going on with fedora
sites? As mirrolist at fedora.redhat.com is not working users are
complaining that the "yum is broken".

Is this why I'm now getting the error during 'yum update' of:

"Cannot find a valid baseurl for repo: core"?

Suggest you look at

Hmmm. That is a condition we have been having on FC6. I hadn't heard it reported on pre FC6. My work around has been
 - make sure the correct repo(s) are enabled
 - yum clean all
 - yum clean metadata
this usually clears the issue.

In some cases I see where they remove the 'cache'.

Yesterday, I went to use k3b on "FC6 pre" and found k3b complained that it couldn't work with the media. It wouldn't work with a DVD-R or a DVD+RW. I wanted to get on with things so I rebooted onto my FC5 hard drive, k3b work great.

Today, after doing a number of trivial things, I decided to have another go at k3b on DVD. But even with the released version of FC6, I still would not use k3b. It does not start to try to erase the DVD+RW, let alone try to burn the DVD+RW.

I changed back from KDE to Gnome and tried "CD/DVD Creator". It had no problem and wrote the image of FC6 on the DVD+RW.

I guess they have broken something on k3b using FC6, because k3b has worked 100% for me up to and including FC5.

I see a bunch of updates for FC6 have been released. I didn't notice anything for k3b, but I will be keeping an open to that subject.

When I burn the DVDs, I must admit I use the 'auto' speed setting. I seem to have the choice under k3b of
   auto, ignore, 1x, 2x

The app seems to try to use 10x to 12x with DVD-R although at some points I have seen it back down to 7x.

For DVD+RW, it seems to try to run at the limit of the media (4x).

I think the system finds it challenging to keep data in the 4 MB buffer. I think that the underrun kicks in sometimes and causes the process to stop while it has a re-go at the process. But when it comes to the verify, the media normally verifies.

On CD, the speed is crazier. The default favours 31x, but I sometimes see it lower that rate to in the 20's.

Both cases (CD or DVD), the access to the hard drive for more source data really works the drive hard to keep data in the buffer. For these higher data rates (ex 17 MB/s), the use of a 4 MB buffer seems marginal.

Re static vs DHCP addr
I used to config for a static addr during the install, knowing that is where I was going to end up. What ever they did abt resolv.conf, I found it looked after itself.

But with these updates and bringing a script from BSD to manipulate the resolv.conf, has forced me to "install with DHCP as default", and when the install is complete, check resolv.conf. The resolv.conf looks good but /etc/resolv.conf/dhclient is empty. I copy the resolv.conf into the dhclient file and then config the network for a static addr.

Just now I checked /etc/resolv.conf* and find the dhclient file has been removed. Something magic must be happening.

Agree with your comment that the ISP service probably limits BW to each IP allowed. Just find it annoying that I do not have more upward BW. Especially after I listed to a couple of Guru's on the TV talk about our facilities in NA vs some of southern asia. We a copper based vs fiber in the Asian area. They didn't have a useful copper installation. Thus the fiber delivers abt 1 order of more BW, in both directions (down and up) to the end user, for abt the same fee as we pay per user per month.

Can't comment on the low latency kernel. I suspect that is true in the FC6. To me, it seems that FC6 is quicker.

Sorry but I won't make TPM. I seem to be low on walking power these days and the doctor is supposed to be looking into it. But I don't see much happening so far. So I will stay quiet. I think I need to excercise more, but will see what the Doc says.

Glenn Simpson VE3DSP
Hamilton, Ont

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]