If you have two unrelated Oracle databases on separate servers, can you replicate A to B and B to A so both servers have a copy of both db's?
Idea is for DR "on the cheap" without buying more Oracle Enterprise licenses. As long as one server could handle the load if required, it seems worth trying. But do Data Guard or other tools let you cross-replicate like this?
We have a three-phase online data center UPS.
Someone said recently that if the input amperage across the phases were 70-10-10 we'd pay for the kWh represented by 70 amps, but if the same load were spread out to 30-30-30, we'd pay for kWh corresponding to only 30 amps.
Really?? I can believe that balancing the load across phases puts less stress on the UPS. But does it also reduce the power bill so dramatically?
We have a new Dell 2950 with PERC 6/e and 14 external SAS 15K 73GB drives. An Oracle 11g database job takes 3 hours to run with the drives set as hardware RAID 10 (striped across 7 mirrored pairs). The database size is about 26GB. The same job running on just two drives in RAID 1 takes only 1 hour. OS is Win 2008 R2.
Before we change the RAID level (with considerable downtime) on the production box, does anyone know why we're seeing this odd result, and if there's a better way to fix it?
ADDED INFO
PERC 6/e should be running the latest firmware and cache battery OK.
FINALLY, THE REAL STORY
After speaking w/the DBA, my face is red. Turns out the RAID 1 is seven RAID 1 volumes of two drives each. The data tables and indices were assigned to each volume to minimize contention. Apparently a good DBA can get more performance from 14 drives than a RAID 10 controller striping blindly across them without regard for file access patterns. Some SANs claim to intelligently migrate files to improve performance, but if there's a bake-off anytime soon, my money's on our DBA!
Our 3560 has a routed interface on the native (untagged) VLAN 1 connected to one port. How would I also pass that VLAN through to another port, but tag it to appear as VLAN 20 on that other port?
EXPLANATION ADDED IN RESPONSE TO COMMENTS:
I have a multi-site Metro Ethernet on the untagged VLAN. Recently added a fiber link from one site to a new site that's not on the Metro. I can just use the native VLAN on the fiber, but I'd like to know if can choose to tag it.
On our Brocade/Foundry FCXs it's easy, and I hoped Cisco had an equivalent:
vlan 20
untagged e 1/1/1
tagged e 1/1/2
You might think of it as a simple form of VLAN translation.
FOLLOW-UP, ADMITTEDLY A STRETCH:
When a physical port sees VLAN 1 (untagged) and VLAN 5 (tagged), can another physical port carry the VLAN 1 traffic but tag it as VLAN 20? In another approach toward a similar end, if port A is an access port in VLAN 1 and port B an access port (untagged) in VLAN 20, can you cross-connect them w/o blowing up the switch, perhaps if you first disable cdp?
ADMT 3.1 copies users with passwords and SID history from my Win2k Domain Controller (native) to a new 2008 R2 DC (R2 functional level) in a new forest/domain.
Can ADMT also copy folders and files from the DC to external DC while preserving a complex set of ACLs, including inherited permissions? Or is there a better way?
If there's a Win2k tool to export ACLs to a file which a Win2008 tool can use to re-create them by matching user names, that might be ideal.
Turns out that the IP of a much-needed new website is blocked from inside our organization's network for reasons that will take weeks to fix. In the meantime, could we set up a reverse proxy on an Internet-based server which will forward SSL traffic and perhaps client IPs to the external site? Load will be light. No need to terminate SSL on the proxy. We may be able to poison DNS so original URL can work.
How do I learn if I need URL rewriting? Squid/apache/nginx/something else? Setup would be fastest on Win 2000, but other OSes are OK if that would help. Simple and quick are good since it's a temporary solution.
Thanks for your thoughts!
We're considering Brocade's CX series vs. Cisco's 3750 at both core and distribution layers for a new site with gig to desktop, without POE.
If you have any hands-on experience with FastIron CX switches, I would greatly value your impressions. I'll add mine to the discussion when we get some quality time with our eval units.
My Impressions:
<Backspace> to work normally. The built-in web GUI looks antique next to Cisco's. Cisco "VLAN trunking" = Brocade "tagging"; Cisco "Etherchannel" = Brocade "trunking."Guess we'll be getting some more CX hands-on!
Routers A, B & C live at 10.1.1.1, 10.1.1.2 and 10.1.1.3 on a /24 metro Ethernet subnet. Each router also has its own private subnet on another interface. Router B's private subnet links thru a firewall to a 10.20.20.0 network at another organization.
Router B redistributes to A and C several static routes for hosts on 10.20.20.0. However, a new host 10.20.20.5/32 must be reached via a different path that goes through router C. I know that C can advertise this host-based route with no problem, but I'd like to keep all my 10.20.20.x static routes in one place.
So, how can B tell A via RIPv2 to send packets for 10.20.20.5/32 to C?
So far it looks like I need no ip split-horizon on router B's 10.1.1.2 interface, perhaps because B has already learned from C other routes with a next hop of 10.1.1.3. But how does RIPv2 split horizon with no auto-summary and network 10.0.0.0 really work? If B learns a route to ANY 10.x.x.x network or host from A or C, is that enough for split horizon to keep it from redistributing ip route 10.20.20.5 255.255.255.255 10.1.1.3?
And if I want to suspend split horizon only for this one new host, how do I filter out the mess of regurgitated routes that B advertises when I try no ip split-horizon?
Thanks much.