I'm not convinced its related but thought I would assk and see
I made a kernel upgrade on six servers to 2.6.14 all of them had 2.4.31 previously now that was the only changes I had made to these systems in the past month or so. Here recently DNS stops, no indication no errors, nothing in logs it just quits. Rebooting the system helps however it doesnt last. When I had asked a friend about it he gave me this
Quote:
This is occuring due to the maximum allowed concurrent dns requests being reached. You can raise the limit by adding the following to your named.conf configuration:
options {
tcp-clients 3500;
recursive-clients 3500;
};
|
Seemed to work but rebooted into an older kernel anyway just to be on safe side. Ive never had this happened before nor had to add this config to named.conf. Heres the confusing part, these six servers have the same exact configuration except the three that are having problems, their dual processors one with HT, why is it that only these three are having the problems and the other three are not?? The other differences in these servers is one has sata drives, other has ide drives, another has SCSI drives but aside from that their all similar, CentOS one with Fedora 1 I used default tcp/ip settings so maybe I forgot something?
This is my kernel config incase anyone wants to see
http://dev1x.100megswebhosting.com/.config
Would appreciate any advice
