gharper's questions

I had a server that was decommissioned & replaced last year, and since the server was no longer in use, I deleted it's client & backup policy from the NetBackup Admin Console shortly afterwards.

I recently got a request to restore a file from the old server, however when I specify the source client for the restore, I get an error message saying:

WARNING: server (backupserver) does not contain any backups for client (oldserver) using the specified policy type (Standard) as requested by client (backupserver).
[Ok]

In addition to that error, I can't seem to run a Client Backup report on the old client any more to determine what tapes I need to recall in order to re-index and restore the files...

Additional info:
- My backup policies are daily incrementals retained for 4 weeks, weekly fulls retained for 2 months, and monthly full backups which are retained indefinitely... I'm trying to restore the monthly backup set.
- All my offsite tapes still appear in the volume pools, however I can't seem to pull up any reports as to what's on the individual tapes, or at least, any tape that was written to prior to August 2009.

My questions:

• Does deleting the client somehow remove NetBackups ability to ever restore files from the old system, even if the backups have a retention period of infinity?
• Is there a way to restore the file from the tape, assuming I can figure out which tape I need?

Running Exchange 2007 on the server with an Outlook 2007 SP2 client, I'm getting an odd message when I try to view specific folders in my Outlook mailbox:

"Cannot display the folder. Your server administrator has limited the number of items you can open simultaneously. Try closing messages you have opened or removing attachments and images from unsent messages you are composing."

I currently have no unsent messages and no emails opened (I'm just trying to look at the list of emails in the folder)... I found someone suggested setting the following registry setting on the server:

HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\MSExchangeIS\ParametersSystem Value: Disable Session Limit

However I don't see that parameter in the registry, and I'm a little hesitant to just add it without knowing if it applies to Exchange 2007 or an older version.

I have mail gateway servers configured to use MailScanner + Postfix + SpamAssassin as described here, along with MailWatch as a web front end.

When sa-learn is run from MailWatch (it's run as the postfix user), it throws this error:

SA Learn: config: path "/root/.spamassassin" is inaccessible: Permission denied, Learned tokens from 0 message(s) (1 message(s) examined)

Running "sudo -u postfix spamassassin --lint -D" gives this bit of info:

dbg: config: read file /etc/mail/spamassassin/mailscanner.cf
warn: config: path "/root/.spamassassin" is inaccessible: Permission denied
dbg: config: mkdir /root/.spamassassin failed: mkdir /root/.spamassassin: Permission denied at /usr/lib/perl5/site_perl/5.8.8/Mail/SpamAssassin.pm line 1577
dbg: config: Permission denied
dbg: config: using "/etc/MailScanner/spam.assassin.prefs.conf" for user prefs file

The bayes tokens are learned correctly, however this error is a minor annoyance and I'd like to fix it... Either by forcing SpamAssassin to not check the /root/.spamassassin/ directory for the config & prefs, or to fix MailWatch so it calls sa-learn correctly & doesn't throw this error.

I'd like to allow certain users to su to another user account without having to know that account's password, but not allow access to any other user account (i.e. root).
For instance, I'd like to allow Tom the DBA to su to the oracle user, but not to the tomcat user or root.

I imagine this could be done with the /etc/sudoers file - is it possible? If so, how?

I have a cluster of Linux servers that are fronted by a load-balancer, which probes each server every 5 seconds to determine if it's operational. These probes generate a lot of useless noise in the logs, and I'd like to simply not have them logged...

Is there a way to prevent rsyslog from logging specific events?

We're getting a few sporadic customer calls (less than 0.1% of our users) complaining of not being able to access my company's website - either they get a blank page (if they use IE), or a "Content Encoding Error" that says the page uses an invalid or unsupported form of compression (if they use FireFox).

The current suspicion is that when an HTTP 1.1 browser request comes through an HTTP 1.0 proxy, we're sending back the compressed HTTP 1.1 browser request, which is somehow being mangled by the HTTP 1.0 proxy.

The website is a Tomcat 4.2.2 back end with an IIS 6.0 front end with GZIP and DEFLATE enabled on the IIS servers.

Has anyone encountered this sort of error before? Is there a recommended fix to avoid breaking older proxy implementations?

Edit: We can replicate the issue with a default setup of squid-2.5, however newer versions of squid seem to work just fine.