[IMPALA-4553] Minicluster starts should wait for NTP to be synchronized before starting Kudu

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: Impala 2.8.0
• Component/s: Infrastructure
• Labels: asf
• Environment: ec2 c4.4xl, ubuntu 14.04
• Assignee: Unassigned
• Reporter: Jim Apple
• Created: 29/Nov/16 17:09
• Updated: 18/Jan/17 22:56
• Resolved: 18/Jan/17 22:56

Description

Sometimes clocks will get unsynchronized and Kudu will not start:

http://35.164.73.121:8080/job/ubuntu-14.04-from-scratch/184/consoleFull

In that job, ntp-wait is called 30 minutes before Kudu starts. By that time:

ntp_gettime() returns code 5 (ERROR)
  time dbe82d4e.8faa68c0  Tue, Nov 29 2016 16:37:34.561, (.561194221),
  maximum error 16000000 us, estimated error 16000000 us, TAI offset 0
ntp_adjtime() returns code 5 (ERROR)
  modes 0x0 (),
  offset 0.000 us, frequency -128.365 ppm, interval 1 s,
  maximum error 16000000 us, estimated error 16000000 us,
  status 0x2041 (PLL,UNSYNC,NANO),
  time constant 6, precision 0.001 us, tolerance 500 ppm,

I've seen this with ?.amazon.pool.ntp.org and ?.ubuntu.pool.ntp.org.

Issue Links

• blocks: IMPALA-3229 Allow all contributors (or at least committers?) to run GVM