seanho
seanho
About
- Username
- seanho
- Joined
- Visits
- 739
- Last Active
- Roles
- Member, OG
- Thanked
- 139
- About Me:
- seanho.com
Comments
-
I assume the client and server can ping each other over the VPN. Firewall on the server? smb.conf / [global] / interfaces? How are clients to login to the samba server?
-
All the best with Syno; I can empathise with wanting set-and-forget usability, especially for your NAS. SHR is basically just btrfs on mdadm. I second the Unraid suggestion; there's a free trial, and it really is very easy to use. Complete flexibil…
-
Congrats Ant on preparing for IH's long-term future and gaining flexibility for your own future endeavours. I can empathise with not wanting to feel stuck in a rut!
-
Catching up here; it seems to me that all these wonderful ideas to lower operating costs of LES would be great topics for a paid admin or paid dev (i.e., someone with more availability than Ant) to investigate. Any takers?
-
Thanks, Ant, for the memoir. Except for the idea to expand NAT LES to help third-world users (which would have been really interesting!), everything else was not a surprise to those of us who've been around LEB/LET for a while. I'd be in favour of …
-
What do you want from your storage? Web interface and WebDAV? Nextcloud. Windows remote mount? Samba over Wireguard / OpenVPN. Fancy distributed fault-tolerant stuff? Gluster, minio, Tahoe, Moose, etc.
-
(Quote) @freerangecloud has regularly-priced VMWare in Vancouver (Harbour Centre, I think?). DC space in western Canada isn't cheap. If Seattle works for you, that may be an alternative.
-
I've had an 0.5TB biennial plan with them since very early days, grandfathered 1Gbps and (slightly) higher iops, still OVZ6 to this day. Only used them for last-resort DR with a remote-dmcrypted loopback device. It's slow and has had short downtime,…
-
I'm running k3s 1.19 on my home lab, yes. I haven't gotten to the point of building a cluster of VPSes yet (but I plan to). On home lab I have 128GB of RAM per node and plenty of disk, and so haven't paid much attention to install size.
-
K3s 1.19 has now reverted to using etcd just like upstream kubernetes. For a long while they didn't have HA, then they tried dqlite, then just gave in to using etcd. You can still use external DB if you like.
-
I agree that k8s is the way forward instead of swarm, but there are still a lot of swarm installations out there. Just a small correction: no pod without a matching tolerance is scheduled on nodes with a given taint. The terminology is super confus…
-
Thanks Tom, I love chatting about editors; I think the UX of software dev is so important! All those little quality-of-life improvements help smooth over the tedious busywork and reduce interruptions to the flow of thought. I was pretty heavily int…
-
Probably best way to do that is via CPU request
-
Excellent pricing for storage!
-
I agree that doing things incrementally and planning on a long-term relationship has the best chance of a positive outcome for the front-line folks that'll be using the system. Baby steps, low-lying fruit, and easy wins. As with most such projects,…
-
2FA via PAM should work just fine in combination with pubkey auth for ssh
-
I think LES is doing well as it is. I don't think there's a need to chase after more traffic, just let it continue to grow organically. Deals will naturally bring in folks. I'm also undecided about more newbie tutorials; DO, Linode, etc have that pr…
-
There's always buster-backports, too. My laptop is running buster but with the backports kernel, in-tree wireguard, managed by systemd-networkd.
-
Debian bullseye also uses kernel 5.7, which includes wireguard in-tree (as of 5.6), so no dkms needed.
-
IIRC they had a similar deal before for .ink, which is another TLD they own.
-
https://www.amazon.com/StarTech-com-Open-Frame-Server-Rack/dp/B00P1RJ9LS How much depth do you have to work with? R210ii, CSE-512, and similar are around 16" rack depth, but most full servers need 26-30" for the rails. Network and A/V rac…
-
(Quote) http://www.orgzly.com/ It's what I'm using now. I'm not in love with it, but it works. I still can't get used to the org-mode way of writing links; I know it predates markdown.
-
SuperMicro 846 or similar 4U, SQ PSUs, swap the case fans for Arctics, use active tower coolers to compensate for the reduced airflow. It's a bit janky but works. I do something similar in a few cheap Rosewill 4U cases; the loudest noise is from the…
-
For DB, my inclination would be to do the HA/failover in the DB, e.g. Postgres streaming WAL replication. Each DB server stores its data on local SSD. You can still use k8s for deployment / lifecycle management of the DB services.
-
All I saw is terrible cable management... ;) I have a 36U rack in my basement with a k8s cluster, five dual-E5v2 nodes plus a couple SFF desktops, 10/40Gb networking. My storage needs are not so big, more compute. My rack is nothing compared with…
-
Unraid doesn't pass TRIM, last I heard. Most folks use it with an array of spinners, plus SSD cache. For 4x NVMe, perhaps zfs pool of mirrors (raid10), depending on your needs.
-
Yes, if you're torrenting a lot it can prematurely wear out both HDDs and consumer SSD. One idea is to get a cheap, used, enterprise 10k SAS HDD (plus HBA) just for torrent/seeding; those are pretty sturdy. Enterprise PCIe/U.2 NVMe can have very hig…
-
LunaNode has load balancers, you just pay $1/mo for the floating IP.