search for: ip_no_pmtu_disc

Ralf Baechle wrote: > I'd like to point those who you need to use these crude workarounds: > > echo 1 > /proc/sys/net/ipv4/ip_no_pmtu_disc > echo 4096 32767 > /proc/sys/net/ipv4/ip_local_port_range > > at a new version of tftp-hpa which solves the PMTU problem by disabling it > only for the tftp client and introduces a new -R begin:end option which > allows to limit the port number range. The changes are about to...

Hi All, I am doing some testing with the asterisk server and have been monitoring the packets exchanged during a SIP-ZAPTEL phone call. I see that the IP-ID in all of the RTP/UDP/IP packets are set to zero. After some googling, I have learnt that some of the linux implementations set the IP-ID to 0 (if the DF bit is set in the IP header) if the two hosts exchanging data are on the same subnet.

...e ws displays "no PXE server found - switching to TFTP:". That is, Etherboot never gets going on ws. I printed your syslinux.zytor.com/hardware.php page where possible problems w/ Intel .99 series is addressed. I am attempting to do the workaround suggested. I set the /proc/sys/net/ipv4/ip_no_pmtu_disc to 1 but don't understand how to disable the blksize on the TFTP server. I have Suse 8.1 as a server. I have their doc but can't find much about TFTP. Please try to refer me if you can't help. Thank you. jr

Hello, I'm experiencing strange behavior on ssh-enabled file transfers (scp and rsync): When connecting to the Internet through an RTC PPP link to a certain provider, file transfers from a remote host (debian potato) to my local host (debian woody) get immediately stuck when issued from the local host, while they work very well when issued from the remote host. On the other hand, no

...ou tweak your MTU sizes? With things like : iptables -A PREROUTING --match policy --pol ipsec --dir in -i eth+ -p tcp -m tcp --tcp-flags SYN,RST SYN -m tcpmss --mss 1361:1536 -j TCPMSS --set-mss 1360 Maybe disable Path MTU discovery to prevent packet fragmentation problems. Helps alo ( net/ipv4/ip_no_pmtu_disc=1 ) Last, login on the proxymod monitor IRQ use, if you have extra cards in the server, you might have one the using a "shared IRQ" I make sure, if possible, that my NIC have there own irq set and not a shared one. And can you post your smb.conf ? That might help also. Greetz,...

...U sizes? No > With things like : > iptables -A PREROUTING --match policy --pol ipsec --dir in -i eth+ -p tcp -m tcp --tcp-flags SYN,RST SYN -m tcpmss --mss 1361:1536 -j TCPMSS --set-mss 1360 > Maybe disable Path MTU discovery to prevent packet fragmentation problems. Helps alo ( net/ipv4/ip_no_pmtu_disc=1 ) > > Last, login on the proxymod monitor IRQ use, if you have extra cards in the server it is a virtual machine on a proxmox root server I just realized that the data transfer rate is going down after a while also on a linux workstation to the same slow speed as on windows workstation....

just to verify basic facts: Did you cross check vie network sniff, on which SMB protocol versions Server + Win 7 clients agree ? Or did you pin down via registry ? AFAIK only starting with win 8 or win 10 clients you could ask with powershell, which protocol version is in use. Did you also cross check samba logs for a name resolution issue ( windows names, not DNS) if one of your boxes is an

I'm trying to network boot a thin client using pxelinux. I had this working about a year ago but have since reinstalled my OS (went from RH9 to FC2). I'm using atftpd. The thin client gets the dhcp address fine and requests pxelinux.0. Eventually, it just times out. Here's the relevant info from /var/log/messages: Oct 7 23:14:36 home atftpd[4717]: Advanced Trivial FTP server started

> Limiting the linux-server's max cpu-speed had > the most affect on performance: > (limited to 1.6GHz instead of 2.4GHz) (33% limitation) The server is equipped with an 6 core / 12 tread 3.6 GHz CPU (Intel Core i7-980x   and  a Raid 10 Disk array. The cpu is 90% idle (including webserver traffic and mailserver traffic) during downloads. There is no encryption because the samba VM

https://bugzilla.netfilter.org/cgi-bin/bugzilla/show_bug.cgi?id=40 laforge@netfilter.org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |ASSIGNED ------- Additional Comments From laforge@netfilter.org 2003-02-03 16:49 ------- We haven't seen this