Comment 6 for bug 805341

I've confirmed this problem with several 10.04 LTS (amd64) machines VM's running on a VMware vSphere system. Each machine that had an uptime of 211 days suffered a sudden softlockup, triggered by an innocent-looking 'apt-get install libyaml-perl'. The machines were running 2.6.32-30-server.

Some consoles showed the "stuck for 17163091968s" message, others just these "stuck for 61s" and "task ... blocked for more than 120 seconds". I've also seen a variant with the value 17163091988s.

Other machines that were quite identical, but running either a bit older kernel (2.6.32-26-server, uptime 365 days) or a newer one (2.6.32-33-server or newer, uptime less than 115 days) were not affected by this issue.

Is there a way to reproduce the error? Maybe a way to set the problematic clock value to somewhere near the 210 day limit?