@matteo-nunziati said in KVM Backing and Support:

@scottalanmiller said in KVM Backing and Support:

@matteo-nunziati said in KVM Backing and Support:

@scottalanmiller said in KVM Backing and Support:

@matteo-nunziati said in KVM Backing and Support:

4. Why do you worry about these things with agent based and not with agentless, even though they are equal and both affected by them just the same?

Because I install altaro have a single admin interface an can backup delta vm in a few minutes.

Sure, but Veeam agentless will do that, too. There are good options in both directions.

Sure. My point was: I'm not aware of a win/linux solution agent based with central management which costs less then agentless. Therefore when I deployed kvm it was with some mix and match stuff got from github and glued into some.bash script. Ok for me but not really nice to offload to others. My ignorance about Good non diy backup solutions is the only reason I do not deploy kvm again.

Grr... Damn phone. Lot of typos...

I’m in the same boat. I’m investigating Bacula, IBM Spectrum, and Commvault… but it seems that Veeam still has an edge.

@scottalanmiller said in Evaluating OneDrive for Business to replace traditional end point backup:

@dustinb3403 said in Evaluating OneDrive for Business to replace traditional end point backup:

@scottalanmiller said in Evaluating OneDrive for Business to replace traditional end point backup:

@obsolesce said in Evaluating OneDrive for Business to replace traditional end point backup:

I'm sure there are cases for it, but not typically as with your own Nextcloud. There, is only one copy of data and no version control or geo redundancy.

NextCloud has version control. And as many copies and as much geo-redundancy as you desire.

But it doesn't include this geo-redundancy with a build it yourself approach. (Usually)

It does every time you choose it.

It's like saying "deploying your own servers don't have passwords." Of course they always do, if you set the password. You can't use "people decide not to do it" for a deploy it yourself solution as if it is lacking it. Everyone chooses how to deploy, everyone has the option.

It would be like saying that most cars you drive yourself can't drive to the beach, but in reality you just meant that most people drive to the grocery store rather than the beach. It's not that the car doesn't go where you want, it's that you chose not to go there with it. Very different things.

ODfB has the amount of georedundancy and protections that it comes with, no more, no less. It has a set amount. NextCloud has as much, or as little, of both as the deployer desires. NextCloud has more protection, optionally, than ODfB.

This is true only to a certain degree. It seems like an exaggeration in my perspesctive. You can make it geo-redundand using external piece of software, of course… like 99% of the software out there.

@dustinb3403 usually, the problem is the LV not mounted at boot time. Check your fstab, try to mount the LV manually etc.

Hi, it’s my business (also)! PM me if you are interested :).

I’m having a good experience with 3CX. Very easy to config and, in cause of a disaster, you can just recover from the single file of backup that it creates with your scheduling.

@scottalanmiller said in Application clustering VS RAID with modern SSD:

@francesco-provino said in Application clustering VS RAID with modern SSD:

Again, valid point. The alternative is to put another pcie ssd in the first node and raid it (mdadm). And, of course, buy another TWO of them and put it in the other node, in case of the first one failed. This is gonna be much higher in costs…

That's misleading because it's not the real alternative. If you were okay with no RAID, but two nodes, you are okay with one node and RAID. Your leap to needing a second node doesn't make sense, it's a level of reliability you don't require. So that's an apple to the orange.

It's a second SSD in the single node and NO second node that is comparable, and easily safer, than two nodes without RAID. Even if it isn't safer, it's REALLY close.

So you can't use the "need a second node with RAID" scenario as a comparison for anything, it's outside of the scope and not roughly comparable. So ignore it, it's not relevant.

Your options are... one node with RAID, or two nodes with network RAID. Single node with regular RAID is faster, simpler, cheaper, and easily comparable if not better for reliability. A second node with no RAID is just vastly impracticable unless it is somehow free while having RAID is not.

Uhm, I’m sorry but I don’t agree with you.
The network RAID will have the same cost (always one other SSD) and BETTER reliability, because even if the mainboard/cpu/ecc fail in the first node, I will have another ready-to-go host with all I need to start my environment.

There is also the possibility of create TWO drbd replica set, one active on the first node and the other active on the second; that way, I can easily double the total cpu count and ram available for the VMs… sort of hyperconvergency on the cheap!

@jmoore especially for 300 users, you’ll better go with hosted PBX. Any internal call will use no WAN bandwidth, simple provisioning, much less bandwidth required anyway, lower costs.

Again, valid point. The alternative is to put another pcie ssd in the first node and raid it (mdadm). And, of course, buy another TWO of them and put it in the other node, in case of the first one failed. This is gonna be much higher in costs…

@guyinpv just put in your cron a daily script that put nextcloud in maintanance mode, run borg in the nextcloud folder (that contains the DB, also) and exit from maintanance mode.

@scottalanmiller said in Application clustering VS RAID with modern SSD:

Cost of Licensing:

Application this is often a costly add on to many software products (and is not always available), and often requires extra software purchases. For example, with MS SQL Server it generally requires more Windows Server and SQL Server licenses, plus additional cost for the application clustering layer. So for many workloads, and any on Windows, the licensing cost soars rapidly.

RAID no known products have any licensing costs tied to block storage redundancy. So there is no cost of this in the real world.

This is true, and I’m trying ti avoid that cost via drbd replication.

@scottalanmiller said in Application clustering VS RAID with modern SSD:

Effort:

Application Clustering requires a lot of expertise, and unique expertise to each and every workload, which must then be monitored, maintained, and updated to keep working. This often triples or quadruples the effort to build and maintain a workload and in extreme cases can be far worse. This is an ongoing effort requiring expertise around maintaining clustering and dealing with edge situations, software changes, and so forth. This is generally outside of the skill set of many IT shops, depending on the workloads. Some clustering, like Windows AD is really simple, some like many databases, is very hard.

RAID zero effort. Tell it to turn on, ignore it. There is nothing to know or do and the system can be safely turned over even to non-technical staff to maintain.

Mostly true, but very different stuff.

@scottalanmiller said in Application clustering VS RAID with modern SSD:

Reliability:

Application Clustering requires everything be duplicated, even CPUs and RAM, so there are some benefits to reliability improvements from the high cost of redudancy. But typically these are minor, as the extra redundancy is typically around pieces that rarely fail. It's a brute force redundancy, rather than a finesse redundancy.

RAID targets the pieces of the system that are most fragile and critical - the storage. It is the drives failing alone that causes full data loss, and drives represent the majority of hardware failures. So you get 99% of the protection, at a fraction of the price.

Because RAID is so mature and reliable, there is an argument that that combined with its insane speed, cache protection options and such will actually be safer than application layer protections that are comparable.

This is unfair, you are really comparing apple to oranges: in one case you have a completely shared-nothing cluster, in the other you are just protected from storage disk failure. What if the cpu/mobo/controller/psu/etc fail?

@scottalanmiller said in Application clustering VS RAID with modern SSD:

Performance:

Application Clustering: Because data has to be synced over the network, there is a performance hit from application clustering. For enough money, you can minimize this greatly, but it just costs more and more to do so.

Traditional RAID: RAID 1 is faster than no-RAID. And moving to things like RAID 10 can speed you up even more. So rather than taking a performance hit, RAID for protection of this nature will speed you up.

Async replication has almost NO performance hit on the master.

@scottalanmiller said in Application clustering VS RAID with modern SSD:

But consider cost and risk of traditional RAID 1 vs. mirrored application clustering for a workload like MariaDB (just as a sample.)

Base Server: $10K

---

Application Clustering: You need two servers, so your cost is $20,000. And that's assuming application clustering is available for the workload, and free. It is free with MariaDB, so this is a good use case.

Traditional RAID: You need an extra SSD for your one server, so say add $500 onto your base cost for a total of $10,500. That's a fraction of the cost of the application clustering.

I already have servers, they are the same spec and out of vendor support.

@scottalanmiller I see your points, but let me just add some additional information about the current configuration:

• we already have three identical server and one NVMe PCIe card;
• I want to use DRBD replication only for stuff that cannot be made high-available without upgrades like SAL server standard etc. Thinking of use syncthing for file replication.

I only use virt-builder… why UEFI?

Thanks @scottalanmiller . I thin I’ll start with a DRBD over two single PCIe NVMe cards (one in each nodes) synced through an infiniband link (infiniband is cheap today!), and I will slowly move every capable workloads to application clustering.

@jblaze said in Latency with VDI in VMware View 7 environment:

We're experiencing latency with our persistent VDIs in our VMware View environment. We're struggling with poor performance with our VDI desktops. Users are experiencing lag when dragging windows between multiple monitors, ‘choppy’ graphics/video, and slow application launching. The problems occur with local users. Both LAN and WAN could be involved. Also, file transfer times from the local desktop to the virtual desktop are super slow. Our bandwidth is dedicated 50Mbps up and down. Is there anything we can do?

Tipically, you will never be happy with that. The only environment I've ever seen where user were HAPPY with VDI was a dedicated LAN with a dedicate Dell m1000 with low per-node desktop density.

@brandon220 I have three home lab servers with Fedora as a KVM host. They are very reliable and fast, IMHO much better than VMware/XS/HyperV. Libvirt CLI is very easy and powerful, I use it almost exclusively to manage the hosts.

You can use cockpit for the only thing that it really can’t do, visualizing the VM vga. You will need it just for fresh installed Windows VM of course, because you can use virsh console (serial) for Linux VMs.

Oh, don’t forget the automation part: I’ve never installed a Linux machine under KVM, you can use virt-builder to get almost every distro (customized with user, ssh keys etc) in seconds. I also automate it with simple script that also use virt-install… pair it with ansible, and you can recreate your environment from scratch with a single command.

@scottalanmiller I’m sure you’ve written something about application clustering somewhere.