Just moved to OnlyOffice Document Server EE (Enterprise Edition). Redis cache is next. If the Radeon Pro v320 can’t be split between multiple VMs, macOS gets dibs. I can’t see myself messing with GPU stuff in Linux again. Getting that to work with nVIDIA drivers was a bit of a pain. By the time I ever try again with another multi-die GPU, it’ll hopefully be RDNA-based and on a server running something newer than Intel Westmere. The show must go on, and one VM can’t absorb all of my time when other tasks await…
Okay, things have definitely taken a different turn than expected. As I learned in the 2021/2022 transition, certain cards that require Above 4G MMIO (Above 4G decoding) are off-limits for me as long as I’m using the DL580 G7. That was part of why I couldn’t use the Tesla K80’s in late 2021. Well, that same situation happened with the Radeon Pro v320, and I have no idea if the same would apply if I had gotten the v340 instead. I got the help of a friend to look for firmware updates, to see if the feature had possibly been introduced in later firmware versions. No such luck. Removing PCIe SSD’s also didn’t help. Sad part is, it looks as though the DL980 G7 didn’t have this limitation. As a result, I will have to push updating the macOS VM to Big Sur further out - until I can get a DL580 Gen8/9 in-house.
I also had difficulty updating software for the secondary DNS (software by Technitium), so I’m going to have to contact them to figure out how to proceed. The handy installer stopped working for me a while back, so I’ve had to perform all updates by hand since mid-2021 iirc.
Hopefully, it won’t be too long before I can get back on track with finishing the rest of the server project, because I miss actually getting tasks done…
Current ToDo's:
- Server/Networking:
- convert the VMs (MBR => GPT, BIOS => UEFI)
Upcoming ToDo's:
- Server/Networking:
- clone SAS HDDs to SAS SSDs
- migrate from ESXi 6.5 to 6.7
Long-term ToDo's:
- Server/Networking:
- VDI host when (DL580 Gen8/9 planning)?
- replace Titan Z with Radeon Pro v320/v340
- macOS Mojave:
- upgrade to Big Sur 11.2.3
- update MacPorts and Xcode
Unconfirmed ToDo's:
- Artix OpenRC:
- Docker container: YaCy Grid
- <https://blog.fossasia.org/creating-a-dockerfile-for-yacy-grid-mcp/>
- <https://github.com/yacy/yacy_grid_mcp/blob/master/docker/all-in-one/docker-compose.yml>
- initiate web crawl
Just kicked FreeSWITCH for izPBX, and actually got izPBX working this time. That was pretty tedious to get working. Now to see if it can survive a container/VM reboot. Last time, it didn’t go so well…
Does Asterisk have AD/LDAP integration? How much work will be required to have it in a basic working config? From what I’ve been seeing online, Asterisk is more of a dev tool than an admin tool. Nothing against it (more flexibility overall), but I’m not sure I’d be able to go at it rn, due to other obligations My time is currently divided.
Looking to see if I can get Cisco CUCM. Might not be likely, but it’s my last attempt at managing a phone system on-prem. Otherwise, I won’t be looking into it again until the next version of this project…
Reorganised ToDo’s to reflect current priorities and project direction:
Current ToDo's:
- Server/Networking:
- convert Windows VMs (MBR => GPT, BIOS => UEFI)
- migrate from vSphere 6.5 to 6.7 (ESXi)
- VDI host when (DL580 Gen8/9 planning) ?
Upcoming ToDo's:
- Server/Networking:
- purchase HPE ProLiant DL580 Gen8
- replace Titan Z with Radeon Pro v340
- Move VMs to new ESXi host (Storage vMotion)
- macOS Mojave:
- upgrade to Big Sur 11.2.3
- update MacPorts and Xcode
- sNTP client configuration (maybe) ?
Long-term ToDo's:
- Artix OpenRC:
- Docker container: YaCy Grid
- https://blog.fossasia.org/creating-a-dockerfile-for-yacy-grid-mcp/
- https://github.com/yacy/yacy_grid_mcp/blob/master/docker/all-in-one/docker-compose.yml
- initiate web crawl
- Install Arch (OpenRC+ZFS on UEFI) from scratch
- move all Docker containers to new Arch host
Yep - planning on possibly moving from Artix to pure Arch, to see if I can bake in the ZFS support I’ve been wanting this whole time. Haven’t found anything (yet) that allows for easy conversion from other filesystems to ZFS, so might be easier to go with ZFS from the beginning in any case. Perhaps My Timeshift backups will be enhanced by this as well…
Still wondering if I should test out a BSD VM one day, once I move to a newer host. Finances are currently very tight, and I still need to purchase a few more EaseUS licenses. The CUCM idea may have to be put off for a while, as I try to figure out the rest of the objectives here. Installing Windows 10 Enterprise on the Threadripper has proven to be more difficult than originally expected. Also considering switching cell service providers in the next 30 days. Life comes at ya fast…
2022 is proving to be a tougher year than the previous one, when it comes to getting major tasks done. I may end up focusing more on the Docker host for a while, since that’s where I’ll be able to make the most progress without breaking the bank. Gotta take a slower pace, to determine a few alternative routes for some of these tasks…
Current ToDo's:
- Cisco CUCM demo/pricing (on hold, due to finances)
- Artix OpenRC:
- Docker container: Discord bridge (matterbridge)
- https://nextcloud.com/blog/bridging-chat-services-in-talk/
- Docker container: YaCy Grid
- https://blog.fossasia.org/creating-a-dockerfile-for-yacy-grid-mcp/
- https://github.com/yacy/yacy_grid_mcp/blob/master/docker/all-in-one/docker-compose.yml
- initiate web crawl
- Docker container: Tor node(s)/relay(s)
- Windows 10 Enterprise:
- convert (MBR => GPT, BIOS => UEFI)
Upcoming ToDo's:
- Windows Server 2016:
- convert (MBR => GPT, BIOS => UEFI) w/ AOMEI license
- Upgrade from 2016 to 2019 (friggin' update times)
- Server/Networking:
- migrate from vSphere 6.5 to 6.7 (ESXi)
Long-term ToDo's:
- Server/Networking:
- purchase HPE ProLiant DL580 Gen8
- VDI host when (DL580 Gen8/9 planning)
- replace Titan Z with Radeon Pro v320/v340
- Move VMs to new ESXi host (Storage vMotion)
- macOS Mojave:
- upgrade to Big Sur 11.2.3
- update MacPorts and Xcode
- sNTP client configuration (maybe) ?
- Install Arch (OpenRC+ZFS on UEFI) from scratch
- move all Docker containers to new Arch host
- or convert existing VM ?!
Finances are still tight, and converting Windows Server from MBR to GPT will cost money since Microsoft left that conversion tool out on Windows Server. AOMIE software license will need to be purchased as a result. Can’t purchase the previously planned EaseUS Backup licenses, due to other expenses that came up recently. I’m at least 2 paychecks (a full month) behind on these acquisitions, and I’m still trying to navigate talks with Cisco, pertaining to CUCM. I’m gonna try to install Windows 10 Enterprise on the Threadripper one last time, sometime in June. Here’s to hoping things lighten up later, maybe in the Autumn…
Boy, things have gotten pretty weird lately. About halfway through May, I learned that I’ll need more licenses for EaseUS (to manage Windows backups). Also had to delay a demo meeting with Cisco, pertaining to CUCM/WebEx. Wanted to add Tor nodes and Matterbridge to the project (as Docker containers), only for the potential risks of the former and complications (between Discord and Nextcloud chat) of the latter to dissuade me on the limited value/returns I’d receive. Days later (05/15), Windows 10 Enterprise refused to install on the Threadripper. Slightly after that (05/19), I find out that Ms left their free MBR-to-GPT conversion tool out of Windows Server 2016 - necessitating the purchase of another software license (AOMIE). On 05/21 (midnight), I moved to Nextcloud 22.2.8, in preparation for the move to version 23. On that same day, the air quality was so bad that I couldn’t run the server later. It stayed off until the 26th iirc. During that time, I re-realised that I’ll need a rackmount AC at some point in the future (had already looked into it before). Between May 27 and 29, I was updating Portainer and purchasing my current domain from Freenom, outright. Also considered hosting an Invidious instance. During the May/June transition, I had to move my mobile devices from StraightTalk to T-Mobile, because the former decided to drop support for the Asus ZenFone 6. They also decided that their VoLTE implementation didn’t need to support said device. And they had a nerve to waste my time (days of it) with poor customer service as I tried and failed to transfer my phone number to T-Mobile (to go easy on people’s contacts lists). Needless to say, Google Voice is the only reason that didn’t end up becoming a complete mess. I stopped giving out my real phone number a while ago, because telephone companies don’t have to play nice. At least T-Mobile has better customer service from what I’ve seen. On June 2nd, I was supposed to look into group policy to reduce Windows 10 telemetry - but got sidestepped by a potential File Explorer bug! On June 4th, I did more research on the requirements for Matterbridge and YaCy Grid. Matterbridge was going to require a ton of work, for something that I’d probably get little-to-no use out of. YaCy Grid is not ready for prime time, and the current config examples only show off Elastisearch. Not to say that they haven’t been working on it, but I’d have to build most of it by hand, which would devour time required for other tasks already planned. The Tor node ended up getting scrapped as well, due to potential risks of hosting said instance. The only thing that survived all of that was the desire to get vanilla Yacy running. On June 7th, I finally managed to give away the Threadripper, freeing up rackspace and the Titan V. On the 10th, I was looking into Docker for Windows containers (no, not that one - the other one with less cringe), for hosting multiple isolated instances of MariaDB. Today, the update to Chromium v102 reigned down on my setup with SSL pains and the terror of frantically scrambling to download/install certificates, while being forced to switch some instances back to HTTP because self-signed certificates are everywhere and sometimes tough to replace. And establishing a proper, self-hosted CA that won’t drain you of your funds is tough. Between all of these, I was battling with tight finances because or a college loan and a certificate loan, and stupidly poor air quality due to temperatures+humidity+allergens preventing me from turning on the server (software and driver updates have to happen sometime). This weekend, I was originally supposed to be 1) configuring group policy for restricting Ms telemetry, and 2) setting up a YaCy instance. It’s 4am. I still have a ToDo list to update!
My time is currently divided.