Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
fping 5.1 (2022-02-06) ====================== ## Bugfixes and other changes - Use setcap to specify specific files in fping.spec (#232, thanks @zdyxry) - Netdata: use host instead name as family label (#226, thanks @k0ste) - Netdata: use formatstring macro PRId64 (#229, thanks @gsnw) - Allow -4 option to be given multiple times (#215, thanks @normanr) - Documentation fix (#208, thanks @timgates42) - Retain privileges until after privileged setsockopt (#200, thanks @simetnicbr) - Set bind to source only when option is set (#198, thanks @dinoex) - Update Azure test pipeline (#197, thanks @gsnw) - Fix getnameinfo not called properly for IPv4 (#227, thanks @aafbsd) - Fixed wrong timestamp under Free- and OpenBSD and macOS (#217, thanks @gsnw) - Documentation updates (#240, thanks @auerswal) - Updated autotools (autoconf 2.71, automake 1.16.5, libtool 2.4.6) fping 5.0 (2020-08-05) ====================== ## Incompatible Changes - In non-quiet loop and count mode, a line is printed for every lost packet (#175, thanks @kbucheli): ``` $ fping -D -c2 8.8.8.8 8.8.8.7 [1596092373.18423] 8.8.8.8 : [0], 64 bytes, 12.8 ms (12.8 avg, 0% loss) [1596092374.18223] 8.8.8.7 : [0], timed out (NaN avg, 100% loss) [1596092374.18424] 8.8.8.8 : [1], 64 bytes, 12.3 ms (12.5 avg, 0% loss) [1596092375.18344] 8.8.8.7 : [1], timed out (NaN avg, 100% loss) 8.8.8.8 : xmt/rcv/%loss = 2/2/0%, min/avg/max = 12.3/12.5/12.8 8.8.8.7 : xmt/rcv/%loss = 2/0/100% ``` - The returned size in bytes now always excludes the IP header, so if before it reported '84 bytes' e.g. when using 'fping -l', now it reports '64 bytes'. This is to make the reported size consistent with ping(8) from iputils and also with fping when pinging a IPv6 host (which never included the IPv6 header size). ## New features - The number of sent pings is only counted when the pings are received or have timed out, ensuring that the loss ratio will be always correct. This makes it possible, for example, to use loop mode (-l) with interval statistics (-Q) and a timeout larger than period, without having the issue that initially some pings would be reported as missing (#193) - Improved precision of measurements from 10us to 1us (#136, thanks @tycho) ## Bugfixes and other changes - The reported size of received packets is now always correct on Linux even for packets > 4096 bytes (#180) - Travis CI automated testing now also macos testing and additional ubuntu distributions (#196)
- Loading branch information