Summary: | Portage 2.0.23 will CRASH your system | ||
---|---|---|---|
Product: | Portage Development | Reporter: | Doug Goldstein (RETIRED) <cardoe> |
Component: | Unclassified | Assignee: | Kevyn Shortell (RETIRED) <trance> |
Status: | VERIFIED FIXED | ||
Severity: | normal | CC: | cardoe, danarmak |
Priority: | High | ||
Version: | unspecified | ||
Hardware: | x86 | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- |
Description
Doug Goldstein (RETIRED)
2002-08-01 06:57:51 UTC
While emerging XFree, any attempt to use emerge -s in another xterm caused the system to hang with 100% cpu and memory use started going through the roof. Was able to kill the second emerge (emerge -s mutt) and system resources dropped back to normal. Using portage-2.0.23, and if it helps emerge -s mutt was still in the dependancies checking stage. Me too. More than 1 emerges cause an endless loop, and an strace has this illuminating data: read(3, "Sandbox error : the SANDBOX_DENY environmental variable should be defined.\n", 4096) = 75 read(3, "Sandbox error : the SANDBOX_PREDICT environmental variable should be defined.\n", 409 6) = 78 read(3, "\33[31;01mACCESS DENIED\33[0m open_rd: /home/jnelson/gentoo-x86/dev-python/snmpy/s nmpy-1.0_alpha4.ebuild\n", 4096) = 104 read(3, "Sandbox error : the SANDBOX_DENY environmental variable should be defined.\n", 4096) = 75 read(3, "Sandbox error : the SANDBOX_PREDICT environmental variable should be defined.\n", 4096) = 78 which goes on forever, sucking up all kinds of memory and eventually killing the systems. ditto here. ditto here. Apparently Azarah has a fix for this and drobbins knows about it... fix in 2.0.24 Cardoe can you verify this as fixed? I do not have a x86 box to test on. Seems fixed here, hasn't happened since around .24. Been working fine since .24... left you a msg on IRC to close this bug a couple of times. Trance... This works! (read previous comment) close this out! Per Cardoe, Fixed |