This is an RFE. Upstream is already planning to overhaul /etc/openshift/resource_limits.conf so I cloned this for tracking purposes. +++ This bug was initially created as a clone of Bug #976607 +++ Description of problem: Setting the memory limit in /etc/openshift/resolve.conf is restricted to memory limit in bytes ('memory_limit_in_bytes' parameter). There is no support to take format '<Limit>' GB or '<Limit> MB'. Version-Release number of selected component (if applicable): rubygem-openshift-origin-node-1.10.1-1.git.130.7e5622d.el6op.noarch Steps to Reproduce: 1. cat /etc/openshift/resource_limits.conf Actual results: [9:51:12 PM] Dan Trainor: [root@qa02node01-3001 openshift]# cat resource_limits.conf # # Limit resources for a new OpenShift guest user on a node # # Standard Profile # node_profile=small # # Quotas: File system limits # quota_files=40000 # 1GB. 1 block = 1024byte quota_blocks=1048576 # Gear usage limits: # "apps" accounting used outdated logic ("git repos" not "gears"). It is being # replaced with "gears" accounting. Logic for max_active_gears falls back to # use max{_active,}_apps if not set, and vice versa. Gears usage/capacity/limits # are counted according to actual gears regardless. # # max_active_gears is used for limiting/guiding gear placement. # For no over-commit, should be (Total System Memory - 1G) / memory_limit_in_bytes max_active_gears=100 # PAM - resource limits # limits_core #<item> can be one of the following: # limits_core=unlimited # limits the core file size (KB) # limits_data=unlimited # max data size (KB) # limits_fsize=unlimited # maximum filesize (KB) # limits_memlock=unlimited # max locked-in-memory address space (KB) # limits_nofile=unlimited # max number of open files # limits_rss=unlimited # max resident set size (KB) # limits_stack=unlimited # max stack size (KB) # limits_cpu=unlimited # max CPU time (MIN) limits_nproc=250 # max number of processes # limits_as=unlimited # address space limit (KB) # limits_maxlogins=unlimited # max number of logins for this user # limits_maxsyslogins=unlimited # max number of logins on the system # limits_priority=unlimited # the priority to run user process with # limits_locks=unlimited # max number of file locks the user can hold # limits_sigpending=unlimited # max number of pending signals # limits_msgqueue=unlimited # max memory used by POSIX message queues (bytes) # limits_nice=19 # max nice priority allowed to raise to values: [-20, 19] # limits_rtprio=19 # max realtime priority # # cgroups - more resource limits # # # cpu # cpu_rt_period_us=100000 # cpu_rt_runtime_us=950000 cpu_shares=128 cpu_cfs_quota_us=30000 # # memory # 1024 * 1024 * 512 = 536870912 = 512MB memory_limit_in_bytes=1536870912 memory_memsw_limit_in_bytes=641728512 # 512M + 100M (100M swap) # memory_soft_limit_in_bytes=-1 # memory_swappiness=60 # # Apache bandwidth limit # apache_bandwidth="all 500000" apache_maxconnection="all 20" apache_bandwidtherror="510" # # Apache rotatelogs tuning rotatelogs_interval=86400 rotatelogs_format="-%Y%m%d-%H%M%S-%Z" Expected results: A parameter to take memory limit in GB or MB Additional info: --- Additional comment from RHEL Product and Program Management on 2013-06-20 22:06:34 EDT --- Since this issue was entered in bugzilla, the release flag has been set to ? to ensure that it is properly evaluated for this release.
We apologize, however, we do not plan to address this report at this time. The majority of our active development is for the v3 version of OpenShift. If you would like for Red Hat to reconsider this decision, please reach out to your support representative. We are very sorry for any inconvenience this may cause.