| Title: | VAX DBMS |
| Notice: | THIS NOTESFILE IS NOT A FORMAL SUPPORT CHANNEL |
| Moderator: | SCARY::CHARLAND |
| Created: | Thu Feb 20 1986 |
| Last Modified: | Tue Jun 03 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 2642 |
| Total number of notes: | 11044 |
Greetings, I have a customer who was seeing lots (1500) of orphaned ruj's when running DBMS V5.1-11. (They have been getting the standard upgrade speech.) Finally, this customer is starting to upgrade to V6.1A (I'm double checking to see if they have eco1). They are quite happy because the number of orphaned ruj's has drastically decreased. They currently have about 20. The question of the day is "Could you ask engineering in what instance these orphaned rujs get generated?" I've explained that they could have been created and the user terminated prior to the root file being updated. I've asked for dumps of the aij's. (Double-checking to see if the ruj's are empty.) Besides a system crash and subsequent database restore, is there any other known reason an orphaned ruj can occur? Thanks, Marilyn
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 2591.1 | HOTRDB::PMEAD | Paul, pmead@us.oracle.com, 719-577-8032 | Fri Jan 24 1997 19:20 | 16 | |
>I've explained that they could have been created and the user terminated
>prior to the root file being updated.
I believe that is the only way known at this time. Assuming that they
have high-water marking enabled on their RUJ disk they should find the
RUJ files are "invalid" (missing proper header record) or do not
contain any data records.
If they do not have high-water marking enabled then it is possible to
have the VMS file system assign a process the uninitialized disk space
that happened to have been used for a previous RUJ file. That can be
confusing since when dumped the RUJ may contain a valid header and have
data records, except it is old and for a different process (PID is
different). This isn't a problem since its filename was never stored
in the rootfile, but it can be alarming to anyone not familiar with
this little "feature".
| |||||
| 2591.2 | M5::MGULLIKS | Marilyn I. Gulliksen - Worldwide Support | Wed Jan 29 1997 13:12 | 6 | |
Thanks Paul. The customer and I are both in class this week. Hopefully, she'll be satisfied the response. If not I'll probably hear about it later this year. Marilyn | |||||