| Title: | ase |
| Moderator: | SMURF::GROSSO |
| Created: | Thu Jul 29 1993 |
| Last Modified: | Fri Jun 06 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 2114 |
| Total number of notes: | 7347 |
Is it possible to set up a single node ASE? I have 4 Rawhides in a memory channel cluster, but we haven't (yet) run ASE. I'd like to use DRD's but not use shared disks. As I understand it, an ASE should be able to serve DRD's to the entire cluster. Of course, I would be happy to put all of the nodes into a single ASE, but I don't want to run a shared SCSI. Can I somehow bring up ASE on the memory channel alone? Alternativly, can I get DRD's running without ASE? It seems like DRD's depend on ASE at some level, but perhaps I can fool the system into making the DRD's work. Any ideas anyone? -drew [Posted by WWW Notes gateway]
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 2059.1 | DRD needs ASE | NNTPD::"pelle@zk3.dec.com" | Pelle Wahlstrom | Thu May 15 1997 14:01 | 24 |
ASE is the only supported way in TCR 1.4 to create and configure drd services. DRD disks MUST be reside on a SCSI bus that is seen exactly the same way by all members of the ASE domain. I don't see any problem in creating a one node ASE domain and create DRD disks on a SCSI bus on that node. However, be aware that ASE will only allow creation of DRD disks if the SCSI adapter supports target mode, which is used for pings between the ASE hosts over the SCSI bus. This is true even if there is only one node in the ASE domain. Basically, this means that you can only use the supported KZPSA (in 1.4) for DRD disks. After the creation of DRD disks you can use drd_mknod (or better drd_ivp) to create drd /dev entries on the other nodes in the cluster. Beware, the ONLY thing you gain with this is remote access to raw disks over the MEMORY CHANNEL. You have NO high availability options (like failover), since there is only one ASE node. If that node reboots or crashes, drd access from ALL the other nodes in the cluster will fail. [Posted by WWW Notes gateway] | |||||
| 2059.2 | Exactly the answer | NNTPD::"kramer@pa.dec.com" | Drew Kramer | Thu May 15 1997 19:38 | 9 |
This is exactly what I was looking for. Do the physical disks have to be on the KZPSA or can I make DRD's out of any disk that is connected to the one node ASE? BTW we don't care about failover, etc... Just trying to distribute the disks across the MEMORY CHANNEL. Thanks again for the info! [Posted by WWW Notes gateway] | |||||