serverfault.com

Have a host of issues, but figured I'd focus on one that I think will at least let me reset IIS. Windows Server 2019 Standard When trying to uninstall the Web-Server Feature it fails, same when I try it via an elevated PowerShell. Uninstall-WindowsFeature -Name Web-Server -Restart Error here Uninstall-WindowsFeature : The request to add or remove features on the specified server failed. Removal of one or more roles, role services, or features failed. The system cannot find the file specified. Error: 0x80070002 At line:1 char:1 + Uninstall-WindowsFeature -Name Web-Server -Restart + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + CategoryInfo : InvalidOperation: (@{Vhd=; Credent...Name=localhost}:PSObject) [Uninstall-WindowsFeature ], Exception + FullyQualifiedErrorId : DISMAPI_Error__Failed_Disabling_Updates,Microsoft.Windows.ServerManager.Commands.RemoveW indowsFeatureCommand Ju

Newbee here that just subscribed to a trial run with Cloud SQL and after having successfully install a DB from phpMyAdmin - ran a plain SELECT query and got this response: "Some of your SQL statements failed to execute. Details: This API does not support reading BLOB columns. Learn more.." and "Cloud SQL Studio doesn't support running SELECT queries on BLOB data. If the SQL script produces a BLOB column in the output, then the SQL script fails with an error." Well, live and learn - What would need to be changed to make this work - or do I have to break it all down with the config and do it over? What instance config supports blob? BR

I currently live in Russia, where internet censorship is increasingly enforced using Deep Packet Inspection (DPI). To bypass this, I’m building a personal VPN setup using V2Ray with the VLESS + WebSocket + TLS protocol stack. As part of my obfuscation strategy, I want to make my traffic look like normal web browsing. I’m running Nginx as a TLS-terminating reverse proxy for V2Ray, and I need the following behavior on the same endpoint /v2ray For regular HTTP requests (e.g. curl -vk https://example.com/v2ray), I want Nginx to serve an HTML page located in /var/www/html. For WebSocket upgrade requests, I want Nginx to proxy them to the local V2Ray service on port localhost:10000. This is needed so that censors see normal web content when inspecting the path, but the VPN tunnel still works over WebSocket.

Using a Fedora 40 (cloud) physical host with podman 5.4.2 I am trying to do multi-arch build. Alpine:3.22.x - amd64 works, arm64 works Debian:trixie - amd64 works, arm64 works Fedora:38,40,42- amd64 works, arm64 will time out Scratch:n/a - amd64 works, arm64 works It appears dnf hangs during DNS resolution. To cover all my bases the same error occures on Fedora 42,40,38. So I tried RHEL UBI10 as dnf and yum are closely related, but also no luck. FROM fedora:40 RUN dnf install grep --verbose Result: Updating and loading repositories: Fedora 42 - aarch64 - Updates 100% | 263.5 KiB/s | 6.4 MiB | 00m25s Fedora 42 openh264 (From Cisco) - aarc 100% | 83.0 B/s | 5.8 KiB | 01m11s Curl error (28): Timeout was reached for https://codecs.fedoraproject.org/op >>> Curl error (28): Timeout was reached for https://codecs.fedo

I would like to send a one-time mass email to several thousand people without getting myself (or, inadvertently, others) labeled a spammer. I'll appreciate advice on the best way to do this. More detail: I am an author with two new books from the same publisher. I want to share this news (and a 20% discount code) with the several thouasnd people who have commented on my blog. How I usually send email: I have an account on a shared server at Bluehost.com, where I host a domain, which I'll call DOMAIN.COM. I send email from a shell account at panix.com, using a "from" line of the form "From: [email protected]" . When I want to mail a couple of dozen people, I write a unix script where each line has the form mail -s "SUBJECT" -f "From: [email protected]" [email protected] < FILENAME This works fine when I am mailing 25 people, but I am concerned about scaling up to thousands. Is there a better way

I'm looking for a reliable way to monitor iptables changes in real-time on Ubuntu. So far, I’ve tried using Falco, which allows me to monitor calls to the iptables binary, but it's not ideal since I won't see the changes made by, for example, iptables-restore. Ideally, I’m looking for a solution that can: Detect all firewall rule changes, regardless of whether they’re made with iptables, iptables-restore, or other tools. Provide visibility into the actual rules that are being changed or applied. Work in near real-time. I’d prefer not to use polling-based solutions, such as a cron job that compares the output of iptables-save every minute. This kind of approach can easily miss quick or transient changes, e.g., if something modifies the rules and reverts them within 10 seconds. Are there better approaches for monitoring iptables changes? Thanks in advance for any insig

We have a kickstart script for Rocky 8 that has worked fine for years on both UEFI and legacy systems. It is failing on newer hardware (Dell Rugged 14) because it seems unable to locate the package repositories, although exactly the same script works just fine on an older Dell rugged laptop. Both systems are UEFI. The script is definitely starting and running through the %pre section; output from there is found in the /tmp directory, as expected, and it looks correct. The repositories are defined, as they have always been, in the kickstart file using the repo statement for our added repositories. AppStream and BaseOS repos are defined in the .treeinfo file, and I can fake their existence by using something like my_BaseOS in the kickstart repo statement, but it fails with or without this. I can see the files are located where I specify in the filesystem, at /mnt/install/repo/<repo_name>, just as spec'd in the repo statements. As noted, this script works fin

We have been experiencing a nasty elusive problem with the GXP1628 IP-phones for several years now. At some moment whole network segment (all devices connected to the same hub) becomes completely unavailable from other segments of the same LAN. The source of that breakdown is one of the phones. When we switch its power off and on again, the problem is solved (for some time). The moment the phone breaks the network is unpredictable: it could be one week or several monthes. The phone is not the same every time (we have about 50), but it seems that the problem tends to apprear on those who have more calls and talk minutes. If we just plug out and plug in again network cable, the problem reappear in a couple of minutes. Software reset does not help, only physical poweroff. We also tried to update phones' firmware. No effect. Last time we managed to isolate broken phone in simple scheme with hub, other phone and notebook with Wireshark (we suspected some kind of flood or abnormal

sftp client closing client after authentication . debug2: channel_input_open_confirmation: channel 0: callback done debug2: channel 0: open confirm rwindow 0 rmax 32768 debug3: send packet: type 1 debug1: channel 0: free: client-session, nchannels 1 debug3: channel 0: status: The following connections are open: #0 client-session (t4 r0 i0/0 o0/0 e[write]/0 fd 4/5/6 sock -1 cc -1) Connection to localhost closed by remote host. Transferred: sent 1852, received 2328 bytes, in 0.0 seconds Bytes per second: sent 38200.5, received 48018.7 debug1: Exit status -1 Connection closed What could be the reason ?

I'm using ModSecurity 3.0.12 version. Nginx is crashing during distributed load testing. Load pattern was 4K users, 6 connections/user, 100M throughput. The nginx worker processes are using more and more memory and eventually crash when the system runs out of memory. The stack trace points to a function used by modsecurity. (gdb) bt #0 0xf7eebdb9 in __kernel_vsyscall () #1 0xf779e257 in raise () from /lib/libc.so.6 #2 0xf779fa93 in abort () from /lib/libc.so.6 #3 0xf76cf90d in __gnu_cxx::__verbose_terminate_handler() () from /lib/libstdc++.so.6 #4 0xf76cd3d3 in ?? () from /lib/libstdc++.so.6 #5 0xf76cd40f in std::terminate() () from /lib/libstdc++.so.6 #6 0xf76cd678 in __cxa_throw () from /lib/libstdc++.so.6 #7 0xf76cdcf6 in operator new(unsigned int) () from /lib/libstdc++.so.6 #8 0xf