Search results

You do not need any activation key for onload, efvi, tcp direct, or ptp with the PLUS version you can check the licensed features once you install it and run "sfkey"

When the heck is the version of ZFS on linux coming out that allows 1. expansion of raidz vdevs, 2. automatic rebalancing of data

This may or may not matter to you, but it irks me to no end....if you add another vdev to your pool all your writes will go to the new vdev, they will not be balanced.

The typical purpose of fronting the ISP connections with a switch is so you can have redundancy with high-availability firewalls. Don't need any vlans, can be unmanaged switches. Each firewall of the HA pair will have a connection to each ISP in this scenario. see this crude diagram edit: If...

The 52 near the end of the model number noted in the auction would indicate 512e

Some are trying other coins that are still PoW, however, that has obvious problems. As miners move to another coin, that lessens the profitability for all miners of that coin. Furthermore, the mining is to support the transactions of that coin...and no coins have anywhere near the volume of...

Uh....environment with a single domain controller? Pretty insane. Having a local backup account isn't a bad idea, however, if you're only creating one to share amongst all your users, you'd have to give it permissions to everything. Everyone would have access to everything. If that's not a...

Even if you have the x10 backplane that supports NVME, you cannot use the M.2 to U.2 adapter for all 10 bays, just 4 of them, which are wired for nvme back to the motherboard. They should work for those 4 bays though.

Yes, please stop calling these arrays, these are not anything like traditional disk arrays that stripe and parity (raid5, raid6, raidz, etc.) I think that's partly why people were not understanding what you were trying to do originally.

Sorry, caught me mid-edit. see my edit2 edit2: what I have described is the method for directly manipulating tape; not using LTFS. I would rather not use an additional layer of abstraction if at all possible; that's just me though LTFS has to do all of these same things under the hood, so...

A file written to tape will be written sequentially. If you're using linux it's very simple to control, read, and write with tape... #rewind tape to beginning mt -f /dev/nst0 rewind #seek to end of data mt -f /dev/nst0 eod #move forward to the beginning of next file mt -f /dev/nst0 fsf 1...

I don't know about proxmox specifically, but it's possible with hardware raid cards, linux md raid, and zfs mirror vdevs. In some of these cases it does warn you about the disparity, and you have to give a flag to force it.

You can still use them together in a raid array, it will just treat all disks as the size of the smallest device.

This idea seems needlessly complex in the extreme. Why not just take the source data and rar/par it if you want to be able to recover from corruption with parity, then commit it to tape in duplicate. Done.

I believe this to be true as well. It appears to simply be a specific configuration of the R630. I manage hundreds of Dell R630's (with some of them being ESXi hosts; zero compatibility issues) and they're great. I wouldn't really recommend a 1U server for homelab though unless you're not...

I'm pleased to announce that with the 9207-8e I'm able to reach 3.7GB/s, which is the limitation of the current disks I have, and not the ds4246 or HBA! Anybody want a 9200-8e? :)

Fragmentation is a layer 3 function, done by routers. If your MTU is too large for layer 2 switching, the frames would simply be dropped. It's fine to leave the MTU at 9000 generally speaking, if your switch supports it. Other traffic, e.g. out to the internet, will be fine since TCP...

Modern systems should not have trouble achieving 2.5Gb/s with standard 1500 byte frames. Typically you start to see issues when it is very small frames, and there are vastly more of them. Anyway, the greater the number of packets, the greater the resources required to process them, primarily...

That's funny you should say that. I was thinking the exact same thing, and was scouring the net to see if anyone had real world benchmarks specifically for the 9200-8e, and I couldn't find anything. Yes, it is old...maybe time for an upgrade to sas-3. However, I was also thinking about...

I'm going to respond to myself after further testing.... I'm still curious what phy36 is exactly, and why it's negotiated at 3Gb/s. I still can't tell if I'm being limited by the # of SAS lanes or negotiated speeds, but these numbers kind of seem to indicate to me that maybe... Here are some...