NAS down, Support Portal not helpful

English General
1

Hi, my nas is down for some days and is not reachable via LAN, only by console. I opended a ticket at QNAP support portal, but got no help for days. First they did not read my ticket and asked nonsens questions. Then they asked for several tests. They got the answers but nobody cares.

Are there other ways to get help? I´m willing to pay a little money but I need my storage and my docker apps back!!!

2

NAS model, setup (backups?) and exact symptoms would be helpful to start with

3

—English Translation added by Dolbyman—
It’s a TVS-675.

Incoming TCP connections from the LAN are failing on all ports (8080, 443, 445, 22) for both wired and wireless clients, including from Home Assistant (wired). Pinging works. A Pi-hole container running on its own IP address exhibits the same behavior. Outbound TCP from the NAS (e.g., curl to 10.0.0.92:8123) works, and a local curl to 127.0.0.1:8080 or 10.0.0.53:8080 returns “200 OK.” After booting, the HDMI display indicates that HD Station is not installed. Access is possible only via the local console (HDMI + USB keyboard, ALT+F2) logged in as admin.

Findings:

  • Services are listening (netstat -tln shows: 0.0.0.0:22, :::8080, :::443, :::445, :::5001); the Recv-Q for all listening sockets is 0.
  • SYN packets from 10.0.0.141 appear in /proc/net/nf_conntrack (as SYN_SENT, [UNREPLIED]), but netstat never shows a state of SYN_RECV or ESTABLISHED.
  • /proc/net/netstat: ListenDrops is increasing (5606 → 6589 → 6632; it rose by 43 during a single browser request); ListenOverflows is 0.
  • iptables INPUT: Policy is set to ACCEPT; the only rule is a DROP for ICMP type 13 (with 0 hits). The mangle and nat tables show nothing unusual; rp_filter is set to 0; routing and ip rule configurations appear normal. - TCP InCsumErrors = 0, standard offloading, no discards at the IP layer.
  • tcp_syncookies = 1, somaxconn = 4096, tcp_max_syn_backlog = 2048;
    sufficient RAM available (approx. 25 GB free/cached), no OOM (Out of Memory) events, no signs of a SYN flood in dmesg.
  • Stopped Container Station and ran network.sh restart: no change.
    Container Station was also stopped and restarted for a data backup: the issue persists.
  • Restarted switch and router: no change.
  • RAID arrays (md1 RAID5 4/4, md2 RAID1 2/2, md3 1/1) and all volumes are healthy and mounted; no fsck processes are running; system load < 1.

Data status: Important data has been backed up to an external USB drive.

4

And you tried to do a 3 second reset?

5

Sounds like an IP address issue. See the pinned post at the top of the forum.

6

Yes, I did. Nas can be pinged but does not answer incoming HTTP, SSH, … and other TCP communications.

7

Hi, which post do you mean?

8

Important to note: this forum allows per-user unpinning of forum topics. What you see as pinned may not be pinned for another user. :wink:

9

I thought it was globally pinned. That was the point…

10
11

Globally pinned means it appears in every sub-forum. But, it can be unpinned by each user when they decide they’ve seen it enough times. :wink:

12

Never knew. Kinda defeats the purpose don’t you think?

13

Kinda.

But, it also allows for user customisation.

Personally, I’d rather acknowledge a globally pinned topic once, then unpin it so I don’t have to see it again. If I think I’ll need it, I’ll save it as a bookmark, or just search the forum for it. I don’t need it thrown in my face every time I visit. :grin: