imPC@ndo IT

Tracker / CVE-2024-41007

CVE-2024-41007

High 7.5

In the Linux kernel, the following vulnerability has been resolved: tcp: avoid too many retransmit packets If a TCP socket is using TCP_USER_TIMEOUT, and the other peer retracted its window to zero, tcp_retransmit_timer() can retransmit a packet every two jiffies (2 ms for HZ=1000), for about 4 minutes after TCP_USER_TIMEOUT has 'expired'. The fix is to make sure tcp_rtx_probe0_timed_out() takes icsk->icsk_user_timeout into account. Before blamed commit, the socket would not timeout after icsk->icsk_user_timeout, but would use standard exponential backoff for the retransmits. Also worth noting that before commit e89688e3e978 ("net: tcp: fix unexcepted socket die when snd_wnd is 0"), the issue would last 2 minutes instead of 4.

Affected products and versions

linux linux_kernel · 4.19 → 5.4.280
linux linux_kernel · 5.11 → 5.15.163
linux linux_kernel · 5.16 → 6.1.100
linux linux_kernel · 5.5 → 5.10.222
linux linux_kernel · 6.2 → 6.6.41
linux linux_kernel · 6.7 → 6.9.10

Analysis

This page is not indexable yet.Until it carries original analysis — what it actually exposes, how to check in two minutes whether a system was touched, what to do if it was — the page stays noindex. The database decides that, not the template.

References