I stuck a pair of Dell 0N743D optics in HPE Aruba 2530 series switches, and for some reason, they worked (after enabling unsupported transceivers in the switches). The slots are SFP, not SFP+, yet somehow SFP+ optics are working there at 1Gb/s rate. Go figure.
what did you learn today? (part 2)
- Thread starter continuum
- Start date
TIL VMware is going to buy Velocloud. My company is actually called out by name (in the quote from Gartner, as an "Enterprise customer") in the press release. 
I fear this means that the web UI (which is really the only UI for the product, and has definitely been flaky and beta-ish at times) will never be completed, given VMware's track record here. :facepalm:
I fear this means that the web UI (which is really the only UI for the product, and has definitely been flaky and beta-ish at times) will never be completed, given VMware's track record here. :facepalm:
I do LAN parties and my network admin from work is involved in them, he finds all sorts of crazy shit on Ebay that we put into the LAN network for shits and giggles.
That to go from VMFS 5 to VMFS 6 requires deleting and re-adding the datastore. One plus though is I can consolidate the numerous small 1-2 TB datastores on the storage.
TIL VMware is going to buy Velocloud.
I fear this means that the web UI (which is really the only UI for the product, and has definitely been flaky and beta-ish at times) will never be completed.
Hey now, VMware has great follow through on finish their clients... ok I can't finish that with a straight face.
TIL VMware is going to buy Velocloud.
I fear this means that the web UI (which is really the only UI for the product, and has definitely been flaky and beta-ish at times) will never be completed.
Hey now, VMware has great follow through on finish their clients... ok I can't finish that with a straight face.
I went back and edited more of that joke into my post before you quoted it...
Plant Ops decided to turn off chilled water at 7:30AM without telling any of the clients of that chilled water. Only had one ESXi host shut down but still...
Is Plant Ops paying for a new server?
They did put out a notice, tho, right?
No notice and I doubt it (on any money) as the host is back up and running now. we just have to suck it up.
I learned that yesterday as well. Shortly after I learned that ESXi 6.5 will happily create a VMFS 5 datastore on a 512e lun despite the fact that it only works in VMFS 6.
Just like how I have to suck it up and continue running on my recently-dried stack switches that I no longer trust, because the CEO doesn't want to bring up the incompetence of building management during lease renewal negotiations.
I once was able to get 4Gbps fibre channel optics working at 1Gbps in a Cisco SG-series switch. Not quite sure how that worked.
Many optics can do both FC and Ethernet.
For example I have a bunch of these Finisar optics on my desk which do 10G Ethernet and 10G FC.
https://www.finisar.com/sites/default/f ... tion_0.pdf
For example I have a bunch of these Finisar optics on my desk which do 10G Ethernet and 10G FC.
https://www.finisar.com/sites/default/f ... tion_0.pdf
Two brand new ASR1001x'es with Cisco branded 1M DAC 10G Cables.
%TRANSCEIVER-3-NOT_SUPPORTE
SIP0/0: Detected for transceiver module in TenGigabitEthernet0/0/0, module disabled
Thanks Cisco. Ugh.
Edit: Turns out my VAR is worthless and sold me an unsupported config. The 1001x'es don't do DAC.
%TRANSCEIVER-3-NOT_SUPPORTE
SIP0/0: Detected for transceiver module in TenGigabitEthernet0/0/0, module disabledThanks Cisco. Ugh.
Edit: Turns out my VAR is worthless and sold me an unsupported config. The 1001x'es don't do DAC.
What's funny is now I'm going to jam some finisar Optics into it until I can get the official Cisco Optics
This thread says they do as of 16.3.1.Two brand new ASR1001x'es with Cisco branded 1M DAC 10G Cables.
%TRANSCEIVER-3-NOT_SUPPORTE
Thanks Cisco. Ugh.
Edit: Turns out my VAR is worthless and sold me an unsupported config. The 1001x'es don't do DAC.
Looks like you need the active cables though, SFP-H10GB-ACU7M, SFP-H10GB-ACU10M
*edit*
Gotta love having to use a 21' cable to connect two devices that are probably 1U away from each other...
Yeah. I'm going to suck it up and grab two 10G-SR optics. Ugh. This is what DAC is for!
This will be the IBGP link between the two (2 ISPs one to each)
This will be the IBGP link between the two (2 ISPs one to each)
This thread says they do as of 16.3.1.Two brand new ASR1001x'es with Cisco branded 1M DAC 10G Cables.
%TRANSCEIVER-3-NOT_SUPPORTE
Thanks Cisco. Ugh.
Edit: Turns out my VAR is worthless and sold me an unsupported config. The 1001x'es don't do DAC.
Looks like you need the active cables though, SFP-H10GB-ACU7M, SFP-H10GB-ACU10M
*edit*
Gotta love having to use a 21' cable to connect two devices that are probably 1U away from each other...
The active cable requirement has bit me a couple of times.
We did find some short active cables that our Brocade was happy with, eventually.
WTF... isn't that the *best possible* time to bring it up? "Listen, you folks caused us preventable losses, how do you make this right?"
We had a formerly wet switch stack at the last job that slowly died one by one over the years.
Found a cool script that looks at the programs that the new "Controlled Folder Access" in Windows 10 has blocked and make it easier to whitelist
https://gist.github.com/gschizas/94f432 ... 4cc7f7a2bb
TIL (again) that some people just don't have a clue or any sense of work ethic.
https://gist.github.com/gschizas/94f432 ... 4cc7f7a2bb
TIL (again) that some people just don't have a clue or any sense of work ethic.
Today I learned that with Sage 100, if a user opens the login screen and doesn't either close the screen back out or actually log in all the way, it'll cause Sage to lock and stop everyone from working until that window is closed or they login all the way.
Does anyone have a good reference for configuring PXE boot options in Linux DHCP? We're planning on replacing our WDS server, and want the new one to support both legacy BIOS and UEFI clients. I have a working setup on our Linux DHCP server now, but only for BIOS clients -- I'm not sure how to add in an option for UEFI. All of the resources I've found so far show how to do it with a Windows DHCP server.
This is what I am using with our ISC DHCP servers.
Code:
class "pxeclients" {
match if substring (option vendor-class-identifier, 0, 9) = "PXEClient";
option tftp-server-name "172.16.0.15";
next-server 172.16.0.15;
if option client-arch = 00:07 {
filename "bootx64.efi";
} else {
filename "pxelinux.0";
}
}Ooh I think we uefi pxe boot some of the sleds in one our products. Let me see if I can dig the config up
Awesome, that's what I was looking for. Wasn't sure if you could do if-then statements in classes, but this is pretty straightforward.
Sage's QA process is little better than Adobe's in my experience. I'm just glad that the customer pays for sage support otherwise I'm not sure I'd have fixed that one as quickly as I did.
And not being a condescending neckbeard to someone isn't brain surgery.
TI(re)L Always put a back up ephemeral port group on your distributed virtual switches.
Creating/editing the bootfile successfully can often times be the worst part of the deal
if you're dealing with the crappy pxe server built into Symantec Ghost Solution Suite.Things can be painful at times.
It used to be that if you had a shitty cluster design (running vcenter on the cluster it managed and having it connected to a distributed VS instead of the standard VS) you could get into a chicken/egg problem without ephemeral ports:
Without an operational vcenter you cannot up ports on a dvs; and if the vcenter runs on a dvs this includes the ports vcenter uses to connect to the hypervisor management interfaces ...
The only way around that was having ephemeral ports configured for the vcenter port groups, but unfortunately the ephemeral ports feature has (or used to have) some pretty serious security implications.
At least that was many moons and vsphere versions ago.
Pretty much. Non ephemeral port groups need vCenter to be up in order to create a port. So if vCenter gets killed somehow and needs to be restarted on another host on the same dvs you are kinda boned.
Here is Chris Wahl's article about it:
http://wahlnetwork.com/2015/01/30/vds-ephemeral-binding/
Here is Chris Wahl's article about it:
http://wahlnetwork.com/2015/01/30/vds-ephemeral-binding/
It is. The consultant that did our initial vCEnter setup before I got hired managed to somehow create a vDS with 1 physical NIC and it worked until we had a blip. I then spent a day building standard switches to mitigate the issue. vDS is great if you need it, but it adds some fault points and additional considerations that might not be worth it for you.