portage-2.2.20 fixes a number of bugs since 2.2.18. There have been no serious bugs reported against this version. This bug depends on Bug 550654 which is already in progress. Reproducible: Always
amd64 stable
x86 stable
portage-2.2.20 seems to gobble memory. portage 14986 0.0 0.0 4192 356 pts/18 SN+ 21:54 0:00 [dev-python/cryptography-0.9.1] sandbox /usr/lib/portage/python3.4/ebuild.sh compile portage 14988 0.0 0.0 123428 408 pts/18 SN+ 21:54 0:00 /bin/bash /usr/lib/portage/python3.4/ebuild.sh compile portage 15005 0.0 0.0 124276 472 pts/18 SN+ 21:54 0:00 /bin/bash /usr/lib/portage/python3.4/ebuild.sh compile portage 15163 19.0 93.7 41117032 15428912 pts/18 DN+ 21:54 2:35 /usr/bin/python3.4 setup.py build root 15326 0.0 0.0 112580 2100 pts/23 S+ 22:08 0:00 grep --color=auto portage 41G of ram before i was able to kill it. have two other machines i haven't been able to kill this process yet. all three servers are updating cryptography
Address Perm Offset Device Inode Size Rss Pss Shared_Clean Shared_Dirty Private_Clean Private_Dirty Referenced Anonymous AnonHugePages Swap KernelPageSize MMUPageSize Locked VmFlagsMapping [...] 7f67f87c8000 r-xp 00000000 08:02 3032733 2128 0 0 0 0 0 0 0 0 0 0 4 4 0 rd ex mr mw me libcrypto.so.1.0.0 0079b000 rw-p 00000000 00:00 0 54836 4828 4828 0 0 0 4828 0 4828 0 49880 4 4 0 rd wr mr mw me ac [heap] 7f67fb423000 r--p 00000000 08:02 2105246 103600 0 0 0 0 0 0 0 0 0 0 4 4 0 rd mr mw me locale-archive 7f5bf8ba0000 rw-p 00000000 00:00 0 50283816 28990200 28990200 0 0 0 28990200 25137436 28990200 0 20117296 4 4 0 rd wr mr mw me ac anonymous mapping, 28G RSS, 50GVSS with 25G of the memory currently referenced.
Stable for PPC64.
(In reply to Blu3 from comment #3) > portage-2.2.20 seems to gobble memory. > > portage 14986 0.0 0.0 4192 356 pts/18 SN+ 21:54 0:00 > [dev-python/cryptography-0.9.1] sandbox /usr/lib/portage/python3.4/ebuild.sh > compile > portage 14988 0.0 0.0 123428 408 pts/18 SN+ 21:54 0:00 /bin/bash > /usr/lib/portage/python3.4/ebuild.sh compile > portage 15005 0.0 0.0 124276 472 pts/18 SN+ 21:54 0:00 /bin/bash > /usr/lib/portage/python3.4/ebuild.sh compile > portage 15163 19.0 93.7 41117032 15428912 pts/18 DN+ 21:54 2:35 > /usr/bin/python3.4 setup.py build > root 15326 0.0 0.0 112580 2100 pts/23 S+ 22:08 0:00 grep > --color=auto portage > > 41G of ram before i was able to kill it. have two other machines i haven't > been able to kill this process yet. all three servers are updating > cryptography This does not seem to be a sys-apps/portage issue, but more of a dev-python/cryptography one, at least with your setups. Please post this issue to a new bug and include your emerge --info. FYI I could not re-produce your issue with either portage or dev-python/cryptograpy. The entire emerge run never used more than about 0.5Gb ram. My 8Gb ram system with desktop and open apps was using ~3.5G of 8GB total, during the run it never went over ~4GB.
bug 554286 created
arm stable
Stable for HPPA.
ia64 stable
Stable on alpha.
ppc stable
sparc stable, closing