IWETHEY v. 0.3.0 | TODO
1,095 registered users | 0 active users | 0 LpH | Statistics
Login | Create New User
IWETHEY Banner

Welcome to IWETHEY!

New yum issues rhel 7
i am getting curl error 56 when trying to download a repomd.xml and a network timout issue is mentioned in the error
using wget it downloads immediately
any ideas?
"Science is the belief in the ignorance of the experts" – Richard Feynman
New One idea
Use wget. :-P
--

Drew
New so you want to come over and spend a few days resolving dependancies? :-)
"Science is the belief in the ignorance of the experts" – Richard Feynman
New Wireshark/tcpdump?
That may at least give you an idea as to where it goes off the rails.
New its not on the network side, monday I will try strace and see what it thinks its doing
"Science is the belief in the ignorance of the experts" – Richard Feynman
New That was my response.
Strace or whatever your local equivalent is.

I find 99% of aborts are internal failures such as attempting to open a missing file, port, pipe, etc, which are not checked or result in a bullshit error message. Days of tech support back and forth or 20 seconds in strace. Always use strace to start.
New caveate, not on a prod box during prod time, strace can knock a box over
I know that you know that, just put it out there for anyone else not familiar with strace
"Science is the belief in the ignorance of the experts" – Richard Feynman
New Actually, I didn't
Well, unless you attach to a kernel process then trigger a deadlock, but I just made that up.

References please.
New here ya go
"Science is the belief in the ignorance of the experts" – Richard Feynman
New You use the term differently than I do
Knock a box over is to cause a panic.
I know strace will dramatically slow the app I am tracing, but it is already aborting and I need to find out why. If it is due to a non timing related issue, then it doesn't matter. If, in a worse case scenario, it causes the target to lock hard, be unable to be run again while the previous instance is locked which means you need to reboot, that would come close, but that is different from an unanticipated panic.
New yes, did not mean cause a panic and a kernel crash but making a system unresponsive to traffic
"Science is the belief in the ignorance of the experts" – Richard Feynman
New fixed but hpfm
took me a while to get back to this but today I had bandwidth
strace gave no clues, on a hunch I changed all of the repo entries from http to https and it started working. Weird because curl and wget had no issue with http but yum would time out.
Don't understand it but it works so on to the next issue.
"Science is the belief in the ignorance of the experts" – Richard Feynman
     yum issues rhel 7 - (boxley) - (11)
         One idea - (drook) - (1)
             so you want to come over and spend a few days resolving dependancies? :-) -NT - (boxley)
         Wireshark/tcpdump? - (scoenye) - (7)
             its not on the network side, monday I will try strace and see what it thinks its doing -NT - (boxley) - (6)
                 That was my response. - (crazy) - (5)
                     caveate, not on a prod box during prod time, strace can knock a box over - (boxley) - (4)
                         Actually, I didn't - (crazy) - (3)
                             here ya go - (boxley) - (2)
                                 You use the term differently than I do - (crazy) - (1)
                                     yes, did not mean cause a panic and a kernel crash but making a system unresponsive to traffic -NT - (boxley)
         fixed but hpfm - (boxley)

We either do it ourselves, or nobody does.
128 ms