Leap Second Adjustments for Linux Servers -
we run number of servers huge system @ work. have been trying manage them leap second added on june 30, 2015. pretty confused plethora of varying instructions found online. instructions differ on different sites, , not yet know 1 follow.
some of things have learnt:
certain version of ntp needs installed , running on servers account leap second.
the kernel version of servers need 2.6.39-400.17.1.el6uek.x86_64 or above.
i not experienced in configuring servers or networking. have been informed kernel upgrades not simple , risky, point might need upgrading new versions of os, in turn might risky our system, in terms of compatibility.
the best guide found online far is: http://www.pythian.com/blog/handling-the-leap-second-linux/
however, while guide explains things better others, says methods not tested, results may vary. means there risk.
i have tried searching lot, , have not found clear easy instructions on how prepare servers event.
can 1 please let me know of reliable/best methods or sources me issue , prepare servers accordingly? system learning management system entire university, extremely critical prepare servers aptly leap second event.
here details of servers based on whatever find best them. not sure if ok , managed leap second or no.
a) 12 application servers following configurations:
os: centos release 6.5
kernel: 2.6.32-431.29.2.el6.x86_64
ntp version: ntp-4.2.6p5-1.el6.centos.x86_64
ntp running: yes
ntpstat message:
synchronised ntp server (10.99.64.42) @ stratum 5
time correct within 700 ms
polling server every 16 s
tzdata version:
tzdata-2014h-1.el6.noarch
/etc/localtime output:
/etc/localtime: timezone data, version 2, 7 gmt time flags, 7 std time flags, no leap seconds, 7 transition times, 7 abbreviation chars
b) 3 synchronized database servers following configurations:
os: centos release 6.2
kernel: 2.6.32-220.el6.x86_64
ntp version: ntp-4.2.4p8-2.el6.centos.x86_64
ntp running: no
ntpstat message:
unable talk ntp daemon. running?
tzdata version:
tzdata-java-2011l-4.el6.noarch tzdata-2011l-4.el6.noarch
/etc/localtime output:
/etc/localtime: timezone data, version 2, 7 gmt time flags, 7 std time flags, no leap seconds, 7 transition times, 7 abbreviation chars
c) 1 nfs server following configurations:
os: red hat enterprise linux server release 6.4 (santiago)
kernel: 2.6.32-358.23.2.el6.x86_64
ntp version: ntp-4.2.4p8-3.el6.x86_64
ntp running: no
ntpstat message:
unable talk ntp daemon. running?
tzdata version:
tzdata-2013g-1.el6.noarch tzdata-java-2013g-1.el6.noarch
/etc/localtime output:
/etc/localtime: timezone data, version 2, 7 gmt time flags, 7 std time flags, no leap seconds, 7 transition times, 7 abbreviation chars
these server infos based on article link. seems knowing information guide 1 find out how ready servers leap second.
kindly please advice me on whether servers ready it, or if not, needs done on servers above them ready leap second. links less confusing sources helpful well. (i tried links given redhat, seem contradict article above , hence confusing well)
looking @ http://lists.centos.org/pipermail/centos-announce/2012-august/018824.html can see kernel fix done in kernel-2.6.32-279.5.2.el6.x86_64 (or later) , means centos 6.4 & centos 6.5 system ok, centos 6.2 need patched/upgraded @ least centos 6.4 handle leapsecond. on other hand - without ntp working no leapsecond info arrive won't issue there ether.
Comments
Post a Comment