serverfault.com

What resources can you recommend on AI and automation? So that I can familiarize myself with them and get advice if necessary. I used Netpeak before.

I am running a small home/SOHO setup with a Synology NAS exported via NFSv3 (sec=sys) and accessed from Linux clients. Current state Linux client: jay -> UID 1000 Synology: jay -> UID 1026 Because NFSv3 relies on numeric UID/GID matching, this causes ownership inconsistencies. I understand that one solution is to align UIDs across systems. In practice that means either: A. Change the Synology user UID from 1026 -> 1000 and recursively chown existing data B. Change the Linux client UID from 1000 -> 1026 Changing the client UID seems invasive, since most local files are owned by 1000. Previously, I manually edited a DSM config file via SSH and ended up in a DSM "mode2 reset" situation. Since then, I am reluctant to modify internal settings outside the supported UI. Questions Which side is generally safer to modify in practice—the Synolo

This Python CGI script gets 3 different results when trying to write a file for each of these 3 directories in Fedora 43: # ll /var/www/cgi-bin/tutorial/test.py -rwxr-xr-x. 1 cpn apache 218 Feb 13 14:29 /var/www/cgi-bin/tutorial/test.py #!/usr/bin/env python open('/tmp/test.txt', 'wt').write('file contents') open('/var/www/cgi-bin/tutorial/test.txt', 'wt').write('file contents') open('/home/cpn/Downloads/httpd/test.txt', 'wt').write('file contents') At the first /tmp/test.txt it does not error but the file is not written. The second /var/www/cgi-bin/tutorial/test.txt is successfully written. The third /home/cpn/Downloads/httpd/test.txt errors with PermissionError: [Errno 13] Permission denied: '/home/cpn/Downloads/httpd/test.txt' Directories permissions: # ll -d /tmp drwxrwxrwt. 28 root root

Wikipedia notes: On rotational drives [data transfer] rate depends on the track location, so it will be higher on the outer zones ... and lower on the inner zones It offers a graph depicting the same - RW of ~120MB/s at outermost of the platter(s), almost half that speed at the innermost. Given, say, a simple mirror of two HDDs, does ZFS automagically write to "opposite" sectors/tracks of each drive, to average out (cancel out, so to speak) these effects on sequential reads back? Or does it write mostly the same to both drives, amplifying the benefits from sequential reads back from the outermost tracks of both drives at once? Can it be configured to balance writes in this way, to flatten the drop-off? Closest answer I could

I have a Fortinet firewall and Active Directory with two Windows Server 2019 servers (DHCP and DNS). There are two ISP (1 and 2) configured for load balancing. There is an external web server connected to ISP 1 with two IP: a public IP for external connections (Internet), and a private IP for internal connections (LAN). Accessing the web site from the internal network I am encountering a potential DNS rebind attack problem. The problem is most likely on the ISP 1 side, but even after many attempts, I have not been able to resolve it with them. I tried routing all internal traffic to the web server to ISP 2 so that the connection would become external, and it seems to be working fine. My problem now is on the internal DNS side: I changed the DNS Forwarder setting of the two Windows Server 2019 servers to a well known public DNS (9.9.9.11-9.9.9.9), the same in the firewall that does not act as DHCP and DNS and is therefore not relevant. Intermittentl

I’m running into a strange issue with a Win32 app in Intune and I’m hoping someone has seen this before. Scenario Win32 app (packaged with IntuneWinAppUtil) Install behavior: User App Installs to:C:\Users<user>\AppData\Local\PDF Buddy Detection rule: Custom PowerShell script Detection is based on a marker file with a version in it: %LOCALAPPDATA%\PDF Buddy\installed.version What happens The install runs successfully. From my install.log: Start install (User) ScriptRoot=C:\windows\IMECache... Detected SourceExe=...\PDF-Buddy-1.1.3.exe TargetExe=C:\Users<user>\AppData\Local\PDF Buddy\PDF-Buddy.exe Copied exe to fixed name: PDF-Buddy.exe Wrote version marker: 1.1.3 Install completed OK. I checked the version marker and the file exists. Then detection runs, and according to my detect.log: === Detection r

I have a server with 2 CPUs(EPYC 9754) on a motherboard(Gigabyte MZ73-LM1), and a GPU(nVidia RTX A6000) attached on PCIE0(of CPU0), and a RAID(HighPoint SSD7505) attached on PCIE1(of CPU1), and each CPU has 64G local memory respectively and a dedicated PCIE bus. This server is being used for deep learnning, so the main working processes are an Postgresql(13) server instance and a few of Python. I'm planning to: configue OS(Debian 12) to bind all interrupt routines of GPU to CPU0; bind all interrupt routines of RAID to CPU1; use local memory allocating policy(a process can only allocate memory that sitting on the same numa node); run all the DB processes on CPU1(near to RAID); run all Python processes on CPU0(near to GPU). My data is not too much, the local memory of single numa node is enough to hold the entire dataset. So I guess that accessing remote-memory(crossing numa node) would only occur a time, i.e

I want to physically move an external USB HDD from one server (raspberry pi 5) to another (also pi 5) with all the data. However, the disk is in a volume group. The disk has only one logical volume on it, so in theory, I can just unplug it and plug it into the other server, and when I do that, the disk spins up fine, but lvm on the new host complains that there is now a duplicate vg and all the lv are inactive. Also the old host complains that it's missing a pv from the vg. How can I move the disk with lv and data on it ?

I have been working to manage a mailing list with mailman3 in a virtual network where I have my own mail server but it always says the same obscure error. This is the content of my /etc/mailman3/mailman.cfg file (I'm only including the section I modified): #... [mta] incoming: mailman.mta.postfix.LMTP outgoing: mailman.mta.deliver.deliver smtp_host: smtp.mydomain.org smtp_port: 587 smtp_user: [email protected] smtp_pass: mailman smtp_secure_mode: starttls smtp_verify_cert: yes smtp_verify_hostname: yes lmtp_host: 127.0.0.1 lmtp_port: 8024 configuration: python:mailman.config.postfix The SMTP host (Postfix, same host as Mailman) works well and a DNS server has a working zone for mydomain.org configured. So Mailman is able to create/delete lists and suscribe/unsuscribe users to/from those lists, but just doesn't send any mail (the notification mails, invitations, etc.). This is the contents of

I have a working postfix installation on a local server, that works by sending my outgoing mail through mailgun SMTP as a smarthost. This setup works. I copied the full /etc/postfix directory on to my remote server, restarted postfix, and outgoing mails don't reach their destination, and get these lines in the logs: Feb 12 07:43:52 mailer postfix/smtp[1480999]: warning: SASL authentication failure: No worthy mechs found Feb 12 07:43:52 mailer postfix/smtp[1480999]: 4BCC4470FB: to=<[email protected]>, relay=smtp.mailgun.org[34.160.157.95]:587, delay=1087, delays=1086/0.05/0.61/0, dsn=4.7.0, status=deferred (SASL authentication failed; cannot authenticate to server smtp.mailgun.org[34.160.157.95]: no mechanism available) So I want to know what went wrong (especially given that when I run a program on the remote server that sends e-mail using mailgun as SMTP server directly, it succeeds): Is there a way to see the SMTP transcrip