rellew Posted April 17, 2019 Share #1 Posted April 17, 2019 (edited) Hi all, I have a strange behaviour with my Microserver Gen8 with 6.2-23739 Update 2. All was working nice and without error for about half a year, when I've shut down the server for holiday. After that, started the server as normally, but when I tried to access the DSM on Port 5000 or 5001 the server was shutting down. The website was showing only the typical Synology log with the message, that the site couldn't be found. I connected with ssh and took a look. When doing the same thing I could see, that the session was aborted with the following message and the server was going down: sh-4.3# Connection to 192.168.178.20 closed by remote host. Connection to 192.168.178.20 closed. So I tried to investigate a little bit and took a look on the error logs of nginx, where I could these lines (/var/log/nginx/error.log) : 2019/04/15 21:24:09 [error] 13240#13240: *34 upstream prematurely closed connection while reading response header from upstream, client: 192.168.178.20, server: _, request: "GET / HTTP/2.0", upstream: "scgi://unix:/run/synoscgi.sock:", host: "192.168.178.15:5001" 2019/04/16 20:02:56 [error] 17294#17294: *1 upstream prematurely closed connection while reading response header from upstream, client: 192.168.178.20, server: _, request: "GET / HTTP/1.1", upstream: "scgi://unix:/run/synoscgi.sock:", host: "192.168.178.15:5000" Further I found these in the /var/log/syslog.log: 2019-04-16T19:40:46+02:00 Skyship syslog-ng[1275]: syslog-ng starting up; version='3.7.3' 2019-04-16T19:40:49+02:00 Skyship syslog-ng[1275]: Connection failed; fd='25', server='AF_UNIX(/var/run/synologan.sock)', local='AF_UNIX(anonymous)', error='No such file or directory (2)' 2019-04-16T19:40:49+02:00 Skyship syslog-ng[1275]: Initiating connection failed, reconnecting; time_reopen='10' 2019-04-16T19:40:49+02:00 Skyship syslog-ng[1275]: Configuration reload request received, reloading configuration; 2019-04-16T19:40:58+02:00 Skyship syslog-ng[1275]: Syslog connection established; fd='31', server='AF_UNIX(/var/run/synologan.sock)', local='AF_UNIX(anonymous)' 2019-04-16T20:02:56+02:00 Skyship syslog-ng[1275]: Syslog connection broken; fd='31', server='AF_UNIX(/var/run/synologan.sock)', time_reopen='10' 2019-04-16T20:03:06+02:00 Skyship syslog-ng[1275]: Connection failed; fd='46', server='AF_UNIX(/var/run/synologan.sock)', local='AF_UNIX(anonymous)', error='No such file or directory (2)' 2019-04-16T20:03:06+02:00 Skyship syslog-ng[1275]: Initiating connection failed, reconnecting; time_reopen='10' 2019-04-16T20:03:16+02:00 Skyship syslog-ng[1275]: Connection failed; fd='29', server='AF_UNIX(/var/run/synologan.sock)', local='AF_UNIX(anonymous)', error='No such file or directory (2)' 2019-04-16T20:03:16+02:00 Skyship syslog-ng[1275]: Initiating connection failed, reconnecting; time_reopen='10' 2019-04-16T20:03:26+02:00 Skyship syslog-ng[1275]: Connection failed; fd='49', server='AF_UNIX(/var/run/synologan.sock)', local='AF_UNIX(anonymous)', error='No such file or directory (2)' 2019-04-16T20:03:26+02:00 Skyship syslog-ng[1275]: Initiating connection failed, reconnecting; time_reopen='10' 2019-04-16T20:03:36+02:00 Skyship syslog-ng[1275]: Connection failed; fd='33', server='AF_UNIX(/var/run/synologan.sock)', local='AF_UNIX(anonymous)', error='No such file or directory (2)' 2019-04-16T20:03:36+02:00 Skyship syslog-ng[1275]: Initiating connection failed, reconnecting; time_reopen='10' 2019-04-16T20:03:43+02:00 Skyship syslog-ng[1275]: syslog-ng shutting down; version='3.7.3' Services are all up and running, data ist still there and as long as I don't try to use the web gui everything seems fine. But it seems the one service, which is faulty, is DSM. When I hard-stop it the server won't shutdown, but yeah...I can't access DSM logically. Hopefully anyone can help me out here. Thanks in advance for this great community! EDIT: I've already tried just to reinstall DSM with the method /etc.defaults/VERSION, but unfortunately when I try to reconfigure and adding the name, IP and password the server is going for a shutdown, because I'm using DSM and it seems I can't reconfigure, when I have the DSM service stopped. Edited April 17, 2019 by rellew More information Quote Link to comment Share on other sites More sharing options...
rellew Posted April 20, 2019 Author Share #2 Posted April 20, 2019 It seems nobody knows about this bug or whatever failure here is happening. After reading a little bit in the forum someone mentioned to reinstall DSM, when you just delete the system partitions. I did this and I can still access my data with with ubuntu & fdisk, but I can't reinstall DSM. When i plug in my usb run the loader (tried reinstall, too) the diskstation can't be found in my network. I pulled the hard disks and bootet once more, but still nothing. Does anyone has a hint here? I just want to reinstall DSM on my disks and keep my data, but unfortunately I can't find my diskstation with the assistent. Quote Link to comment Share on other sites More sharing options...
rellew Posted April 20, 2019 Author Share #3 Posted April 20, 2019 It seems the usb flash drive was faulty. Did a new one with Jun's loader 1.03, network was working again with it and I could migrate my data after that. Gave me alot of trouble for nothing - if you should stumble over this thread, just make a backup of your config, create a new flash drive and install DSM over it to keep your data. Quote Link to comment Share on other sites More sharing options...
naasking Posted March 4, 2020 Share #4 Posted March 4, 2020 I'm running into exactly this error. Had DSM 6.2 running fine on this exact same hardware to backup my old server, then swapped the old server disks into this machine and migrated. Everything seemed fine except accessing DSM and ssh simultaneously kills the network and all network services, and I see these syslog errors everywhere. I've tried a reinstall with a different USB drive and the same flash image, same problem. I just tried a brand new flash drive with a freshly constructed flash image, as I thought perhaps the image itself got corrupted, but the problem remains. I think at this point whatever the problem is, it must be on the system partitions and it must somehow persist across reinstalls. Unless anyone has any other suggestions, I might have to do a completely fresh reinstall instead of trying to preserve all of my settings. Quote Link to comment Share on other sites More sharing options...
naasking Posted March 5, 2020 Share #5 Posted March 5, 2020 Well, a fresh install seems to have done the trick. My previous configuration had been updated since DSM 5.1ish, so some errors had to have crept in over time. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.