> I have noticed that Anaconda may stall when installing RPMs over a > LAN connected to a hub. What sort of problems can cause Anaconda to > timeout completely? > > How can I diagnose these problems? > Not that this is your problem you are experiencing, but anaconda depends on rpm to a great degree so things that would cause rpm to stall would cause anaconda to stall. There are two "general" cases that I have seen rpm stall (though not necessarily in anaconda). One is an rpmdb lockup, the other is a scriptlet lockup (much more rarer btw). In both cases if you ran strace against rpm you would see rpm waiting on a futex (or rather it is in the FUTEX syscall). You can determine which one by running gdb against rpm and getting a backtrace. In the scriptlet lockup you have passed through runScript() in the stack at one point, with db lockups your into the db code. That said neither of these may be your lockup. Cheers...james
