Bug 1493069

Previously, Red Hat has updated the default values in
/usr/lib/sysctl.d/00-systen.conf for architectures other than s390.

Few months ago, we received a BZ report #1458916, that the default values for s390 were kept as they were before, and which are too low for some applications to run correctly.

Therefore, I would like to hereby ask IBM representatives, what should be correct default values on IBM's s390 (and s390x if needed) architectures for these settigs?

----------------------------------

# Kernel sysctl configuration file
#
# For binary values, 0 is disabled, 1 is enabled.  See sysctl(8) and
# sysctl.conf(5) for more details.

# Controls the use of TCP syncookies
net.ipv4.tcp_syncookies = 1

# Do not spend as much process time too early to write out dirty pages
vm.dirty_ratio = 40

# Increase the average time a process runs continuously and also improve the
# cache utilization and server style workload throughput at minor latency cost
kernel.sched_min_granularity_ns = 10000000
kernel.sched_wakeup_granularity_ns = 15000000
kernel.sched_tunable_scaling = 0
kernel.sched_latency_ns = 80000000

# Controls the maximum shared segment size, in bytes
kernel.shmmax = 4294967295

# Controls the maximum number of shared memory segments, in pages
kernel.shmall = 268435456

----------------------------------

(https://github.com/fedora-sysv/initscripts/blob/rhel7-branch/sysctl.conf.s390)

----------------------------------

Would IBM be okay with completely removing all of these settings, and using default kernel values?

Would the current config (used for all other architectures) https://github.com/fedora-sysv/initscripts/blob/rhel7-branch/sysctl.conf be sufficient for the needs of IBM s390 (and s390x) arch?

Thank you in advance!

----------------------------------

@dhorak: Please, forward this request to IBM. Quick response from IBM is welcomed.