On Tue, Dec 2, 2008 at 10:10, Paul Black <paul+fedora@xxxxxxxxxxxxxxxx> wrote:
2008/12/2 Andre Costa wrote:Actually, is this a problem at all or can I safely ignore this?
As Wireshark suggested, this is happens the NIC calculates the checksums for the packet. Since the HW is doing this, the SW doesn't bother, hence the wrong value that appears to be in the outgoing packet. You would need to verify at the receiving end whether there was an actual problem but this is unlikely to be the case.
Thks, so it's probably nothing to worry about.
I digged even further and found out that Google-related queries (GMail, GReader etc.) are returning very short TTL values, as can be seen by the wireshark capture snippet below:
Answers
www.google.com: type CNAME, class IN, cname www.l.google.com
Name: www.google.com
Type: CNAME (Canonical name for an alias)
Class: IN (0x0001)
Time to live: 10 hours, 31 minutes, 12 seconds
Data length: 18
Primary name: www.l.google.com
www.l.google.com: type A, class IN, addr 74.125.77.99
Name: www.l.google.com
Type: A (Host address)
Class: IN (0x0001)
Time to live: 1 minute, 50 seconds
Data length: 4
Addr: 74.125.77.99
www.l.google.com: type A, class IN, addr 74.125.77.147
Name: www.l.google.com
Type: A (Host address)
Class: IN (0x0001)
Time to live: 1 minute, 50 seconds
Data length: 4
Addr: 74.125.77.147
[...]
(first one is www.google.com, which has a TTL of 1 day)
I bet this is why no local DNS cache can help with Google addresses since they expire so quickly. Weird thing is that, on this short monitoring I did (a couple of minutes only), none of the Google sites (aside from the main sites such as www.google.com and mail.google.com) had a TTL > 5min. Has this always been this way?
Regards,
Andre
-- fedora-list mailing list fedora-list@xxxxxxxxxx To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list Guidelines: http://fedoraproject.org/wiki/Communicate/MailingListGuidelines
