(Would this be better on MVS-OE?)
This is utterly weird! On Solaris 10, I do mvslogin, then cd to
a PDS, then do "wc *" (count lines, words, and characters in all
the members). I see:
135$ ( cd /my/mvs/PDS && wc * )
0 0 0 a
0 0 0 b
0 0 0 c
0 0 0 d
0 0 0 e
^C
139$ (stopped)
The members appear to be there, but empty!
At about the same time, on a Solaris 8 system, I do
the same and see:
137$ ( cd /my/mvs/PDS && wc * )
0 0 0 a
0 0 0 b
0 0 0 c
0 0 0 d
0 0 0 e
0 0 0 f
59 301 4779 g
143 1129 11583 h
138 1096 11178 i
96 770 7776 j
^C
139$
... the members that Solaris 10 reported as empty appear likewise
empty to Solaris 8! Members beyond that point appear OK. If I let
Solaris 10 run longer, more members appear empty on Solaris 8. If
I wait long enough and try again on Solaris 8, everything looks fine.
Surmise: when I attempt to read a member with Solaris 10, something
goes wrong; the member is reported as empty. But something caches
this information (who else but MVS.MVSNFS?)! So members read
incorrectly on Solaris 10 appear likewise incorrect on Solaris 8
until the cache decays.
Has anyone else seen anything like this?
Ugh!
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
