lukecyca's questions

I have a 10TB XFS filesystem on a SLES 10.2 server. I am currently running rsync on it once per day, to sync it with a disk-based online backup on another server. The primary volume is quite busy as it is, and so rsync is only running at 2MB/sec on average. However, I can scp a file and get >50MB/sec to the same destination. This leads me to believe that rsync is overburdened with deciding what to transfer, not actually transfering.

What other ways can I achieve this type of synthetic full backup? What are the relative pros/cons of the following ideas:

1. Use xfsdump, piped through ssh or similar, to xfsrestore on the destination. Can it do incrementals when used in this way?

2. Some of sort of asyncronous block replication using drbd or similar.

3. Some better use of rsync such as looser comparison criteria. I'm currently using

   rsync -aqAX --numeric-ids --delete-during --files-from=foo ...

I have an 8-core linux-based server dedicated to serving NFS to 80 linux clients that run batch jobs. The clients have 400 cores in total, and so are generally running 400 single-core batch jobs concurrently.

Occasionally, lots of batch jobs try to do I/O at the same time, and exhaust the number of nfsd threads on the server, of which there are currently 80. The batch job gets an I/O error (such as Permission Denied) and bails out.

I'd like to increase the number of nfsd threads, but want to know:

• What rules of thumb are there for setting the number of threads in this situation?
• What drawbacks are there to setting it too high?

References

• This NFS tuning guide from Sun suggests some rules of thumb for Solaris, but gives no rationale for those particular numbers, so I'm not sure how they apply to my Linux server.
• This other one gives an approach to this type of tuning, but is highly subjective.

I want a high-level test that ensures that both incoming and outgoing email services are operating normally. I've achieved this by writing a python script which does this:

1. Send a message with a unique hash through my mail server to [email protected]
2. The gmail account is configured to auto-reply back to the same address and then delete the message.
3. My script polls IMAP until it finds an email with the correct hash sitting in its inbox, or times out.
4. It reports the elapsed time to my monitoring software (Zabbix)

My question: Is gmail the best third-party to use? Should I add a couple others as well such as hotmail and yahoo? Is there anyone more official that will auto-reply to these sorts of "mail pings"? Any other recommendations for this type of test?

I have about 120 DLTape-VS1 tapes and 2 drives that are all near the end of their retention period, and have been superseded by newer technology. I would like to donate the whole package to my local FreeGeek organization in the hopes that someone else can get some more use out of it (many of the tapes have only been written once or twice). But first, I need a way to erase all of the tapes so that my company's data isn't leaked.

What's easiest/cheapest?

• Build some sort of machine with both drives and manually blank each tape? How?

• Use some sort of fancy magnet thing to passively wipe the data away?

• Any other creative ways to reuse/recycle this investment in an ecologically friendly way?

I want to enable encryption on all of my backup tapes. I more-or-less know how to do this technically, but the procedural and human elements of implementing this are tricky.

I use HP LTO4 drives with bacula, which doesn't have any key-management features. In fact, its support for hardware encryption is to call an external script which sets the key on the drive before reading and writing.

My questions:

1. How should I keep track of which tapes have encryption? I already have a few hundred tapes without encryption. Even if I take the time to rewrite them all with encryption, there will be months of overlap where some have it and some don't. How will bacula know whether to set the key before reading a given tape? Is the drive smart enough to read unencrypted tapes even when a key is set?
2. If the key is ever compromised, we'll have to change it and we'll have the same problem as #1.
3. If the key is lost, we've effectively lost all of our backups. How can I mitigate this without increasing the risk that it is compromised?
4. Should the key change regularly? Once per year? What is the best practice?
5. How do the big ISV backup systems handle these issues?

I am migrating data from one S3 account (billed to me) to another S3 account (billed to my company). How do I copy the data in the easiest, fastest, and most cost-effective way?

There are some utilities that do this (like S3 Browser), but they all seem to be Windows-only (no good for me), and I don't really want to buy a piece of software to do a one-time copy operation.

Thanks!