Thanks to the list for the many responses I got in so short a time.
My original post is at the bottom.
The replies to my post varied only on a couple of points
1 - All agreed that I should not need any firmware patches
2 - Most were of the opinion that 2.7 was y2k ready as long is it had the
recommended patches. A patch cluster was mentioned but I could not find
one. I did find a couple recommended patches (they dealt with lastlogin
and C compiling) and they've been applied.
3 - Burning the time into the system board was NOT mentioned as anything
to be concerned about.
4 - The answers were split on this one. Many said I should simply restore
the whole box from scratch via tape backup, and many said that I wouldn't
need to.
5 - All agreed that I should turn off ntp, many said that cron would die or be
affected, as well as anything that processes log files. Many recommended that
I simply turn off my ethernet interfaces. And many recommended that I do all
of this in single user mode.
What I will more than likely do after I turn the clock back is to try a
find -exec touch, to put the files with future dates back to the present.
If that doesn't work, or if anything is amiss or just wrong, I will have my
backups at the ready.
I have a much clearer picture of how I wish to do the y2k testing now.
Thanks a lot for the input. Timely as usual.
Rich
Original post follows:
---------------------------------------------------------------------------
Hi,
I am going to do some y2k testing on my firewall.
It is a sun sparc ultra 5 running solaris 2.7.
I have read through the archives specifically regarding y2k
testing/rollback.
The most recent article being Feb. 99.
I am looking for feedback regarding a few assumptions, namely that:
1 - I shouldn't need any firmware patches for the ultra 5
2 - The sun site(last time I checked) did not have any y2k patch clusters
for 2.7, but did recommend individual patches: 107587-01, and
107359-02.
3 - After setting the time ahead, I should not have to worry about burning
the new time into the system board.
4 - Rolling the machine's time back is okay, but it is better to simply do
a full tape restore(but not necessary?).
5 - I'll want to consider turning off the ntpd, and email during the
testing.
My question is:
A - Are these assumptions on or off base?, and
B - Did I miss anything significant?
I welcome any relevant type of y2k testing horror stories.
and thanks in advance for any/all responses and I will
summarize as this seems to be one of the more widely
relevant topics,
Rich
--------------------------------------------------------------------------
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:13:32 CDT