Re: [OpenAFS] ADS and MIT Kerberos transition auth continued

2009-07-09 Thread Vincent Fox
Simon Wilkinson wrote: On 9 Jul 2009, at 16:50, Douglas E. Engert wrote: Depends on what data you put in AFS, and is the AFS network traffic sniffable You would need to do a risk assessment of you situation. And when you do that risk assessment, consider the sentiments expressed in: http

Re: [OpenAFS] ADS and MIT Kerberos transition auth continued

2009-07-09 Thread Simon Wilkinson
On 9 Jul 2009, at 16:50, Douglas E. Engert wrote: Depends on what data you put in AFS, and is the AFS network traffic sniffable You would need to do a risk assessment of you situation. And when you do that risk assessment, consider the sentiments expressed in: http://xkcd.com/538/ S.

Re: [OpenAFS] ADS and MIT Kerberos transition auth continued

2009-07-09 Thread Douglas E. Engert
Eric Chris Garrison wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Jeffrey Altman wrote: Garrison, Eric C wrote: 07/08/09 14:53:56 07/09/09 00:53:44 afs/afstest.iu@ads.iu.edu renew until 07/09/09 14:53:40, Etype (skey, tkt): AES-256 CTS mode with 96-bit SHA-1 HMAC, AES-256

Re: [OpenAFS] ADS and MIT Kerberos transition auth continued

2009-07-09 Thread Russ Allbery
Eric Chris Garrison writes: > Jeffrey Altman wrote: >> The answer is right above. AES-256 is not DES-CBC-CRC > > I'm told by our ADS admin that DES3 isn't supported, That wouldn't help; AFS doesn't support DES3 anyway. > and DES-CBC-CRC is somewhat weak by modern standards. How concerned > sh

Re: [OpenAFS] ADS and MIT Kerberos transition auth continued

2009-07-09 Thread Eric Chris Garrison
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Jeffrey Altman wrote: > Garrison, Eric C wrote: > >> 07/08/09 14:53:56 07/09/09 00:53:44 afs/afstest.iu@ads.iu.edu >> renew until 07/09/09 14:53:40, Etype (skey, tkt): AES-256 CTS mode >> with 96-bit >> SHA-1 HMAC, AES-256 CTS mode with 96

Re: [OpenAFS] Network becomes terribly slow when cache manager flushes updates over xDSL

2009-07-09 Thread Harald Barth
> The standard is ICMP WOULD FRAGMENT. We have tried years ago. MTU detection through ICMP is not feasible on networks you do not control. > People who block that don't deserve functional apps. Don't cater to them. Unfortunately the people who block are not the ones who take the hit. The amoun