Mailing List virtualized_ecs_users@2rosenthals.com Archived Message #208

Fra:	"Lewis G Rosenthal" <virtualized_ecs_users@2rosenthals.com>	Full Headers Undecoded message
Emne:	Re: [Virtualized eCS] To upgrade to VBox 7.1.6 or not
Dato:	Wed, 16 Apr 2025 17:41:06 -0400
Til:	Virtualized eCS Users Mailing List <virtualized_ecs_users@2rosenthals.com>

Hi, guys... On 04/16/25 04:35 pm, Doug Bissett wrote: On 2025-04-15, at 14:28:45, Allen Heath wrote: I tried switching to the Intel Pro/1000 MT Desktop & IBMGU driver and I still get the same "out of resources"results as with what I have been using for about 3 years with the PCnet-Fast III adapter. I can casually copy lots of stuff back and forth to a remote box if I don't do too much too quickly, otherwise I get the "network out of resources" message. And simple things such as just a "dir" on the remote drive will then fail ... for a while. However if I walk away for about 10 minutes the situation cures itself and operations to/from the remote drive work again. And then if I copy a lot of stuff again it will run out of resources once again. Have you tried copying say a 600,000,000 byte image file and running a md5 checksum on it for comparison? I might get two "loops' of doing that before I have to wait 10 minutes... and resume/retry. If just doing small files spaced out during the day, no problems. I just copied a 695MB (ZIP) file, running MD5 against it, as quickly as I could, 6 times with no problem. I think you have some other problem. Whether it is VBOX (including your settings), the nic and driver (unlikely), or win 11, I don't know. If this were an ArcaOS guest behaving like this, we would ask for a testlog of the system right after bootup, a restart of the VM, and another testlog taken after the symptom would present. While the TestLog application itself will run on eCS (even Warp 4, post-FP5 or so), opening a ticket for such analysis would not be something we would do as eCS isn't our OS, and we don't provide such support. All of that having been said, it is definitely not normal for the VM to run out of resources unless something else is at play, likely within the VM itself (some other software or outdated component). We would do the analysis if the problem surfaced only after installing one of our NIC drivers available by subscription, in case the situation was exposing a bug in one of our drivers (not likely in this case, or we would see it on bare metal, as well). By the way, ftp or curl etc experience the same out of resources if I do too much too fast, but also cures itself after a significant pause. This is very odd. Perhaps. Garbage collection can take some time, depending on a number of factors, and we have no information as to the configuration of the virtualized hardware or the OS (CONFIG.SYS, etc.). Without that data, all of this is pure speculation. Clearly, something is not right, and this problem is most unusual under VBox with a typical eCS setup (but what's "typical," I wonder?). I haven't tried those scenarios with my Ubuntu guest. That guest was only installed as a learning experience, but I'm still not a Linux fan. If I get a significant break on what's going on in my life I might experiment with the Linux guest to seem if it is common problem with my Host. Another option I might experiment with is running an image copy of my Host as a guest of itself. I heard someplace it was possible to run another virtualization under a virtualization. That would be to simply test if the Vbox 7 upgrade breaks my otherwise working host. Three or so years ago I could never get a 7.x version to install successfully. I wouldn't waste my time doing that. It may, or may not, work, depending on what the host machine supports. Running a windows guest is nothing, at all, like running an OS/2 guest. Yeah, nested virtualization is fun to ponder, but not so much fun in reality. On PCs, it never works like it does on a VAX, where you can seemingly endlessly nest VMs. PCs just weren't designed for such things, and other problems creep in one, two, three layers down. The processing overhead can get to be quite burdensome and can be quite palpable, visible as lack of responsiveness, slow screen updates, poor network performance, and so forth. It's simply not worth the effort of setting up. For debugging something like this, I would start over with a fresh VM and a vanilla configuration. Test from that with a minimal set of ancillary packages installed and running, and gradually add more of your desired/usual processes to the configuration until you find what is triggering the bad behavior. -- Lewis ------------------------------------------------------------- Lewis G Rosenthal, CNA, CLP, CLE, CWTS, EA Rosenthal & Rosenthal, LLC www.2rosenthals.com visit my IT blog www.2rosenthals.net/wordpress -------------------------------------------------------------

Abboner: Feed, Digest, Index.
Stopp abbonement
E-post til ListMaster