Too may redirections (599) pve4

[RobFantini]

i followed wiki and deleted node from pve3 cluster.

then did fresh install of pve4 .

used same hostname and IP .

then added to pve4 cluster.

at pve web I get ' Too may redirections (599) ' display when i click on a vm or try to do a restore .


syslog show a lot of this :

Oct 12 12:02:00 sys5 pveproxy[3752]: problem with client 10.1.10.181; rsa_padding_check_pkcs1_type_1: block type is not 01
Oct 12 12:02:00 sys5 pveproxy[3752]: problem with client 10.1.10.55; rsa_eay_public_decrypt: padding check failed
Oct 12 12:02:00 sys5 pveproxy[3752]: Can't call method "timeout_reset" on an undefined value at /usr/share/perl5/PVE/HTTPServer.pm line 225.
Oct 12 12:02:00 sys5 pveproxy[3751]: EV: error in callback (ignoring): Can't call method "push_write" on an undefined value at /usr/share/perl5/PVE/HTTPServer.pm line 295.
Oct 12 12:02:01 sys5 pveproxy[4924]: problem with client 10.1.10.181; rsa_padding_check_pkcs1_type_1: block type is not 01
Oct 12 12:02:01 sys5 CRON[5168]: (root) CMD (/fbc/bin/fbc-sleep-300 ; /fbc/bin/pve/pve-fbc-misc.sh   >>/var/log/pve-fbc.log # do b4 log rotate)
Oct 12 12:02:01 sys5 pveproxy[4924]: EV: error in callback (ignoring): Can't call method "push_write" on an undefined value at /usr/share/perl5/PVE/HTTPServer.pm line 295.
Oct 12 12:02:01 sys5 rsyslogd: [origin software="rsyslogd" swVersion="8.4.2" x-pid="3580" x-info="http://www.rsyslog.com"] exiting on signal 15.

any clues ?

I did start a migration then cancelled.. then completed after removing an unused disk...

 

[wolfgang]

Hi,

you can't mix pve3 and pve4 in one cluster.

 

[RobFantini]

Hi,

you can't mix pve3 and pve4 in one cluster.

sys5 was added to the new cluster. it shows red on old cluster web page,

new cluster:

sys5  ~ # pvecm nodes

Membership information
----------------------
    Nodeid      Votes Name
         4          1 sys5 (local)
         2          1 sys4
         1          1 sys3
         3          1 pve126

old:

# pvecm nodes
Node  Sts   Inc   Joined               Name
   1   M   4700   2015-10-10 19:51:16  dell2
   4   M   4704   2015-10-10 19:51:17  sys7
   5   M   4480   2015-10-10 09:06:21  dell1

The warnings stopped an hour after I posted to forum.
I was able to add vm to sys5 after the warning stopped.

On old cluster web gui:sys5 is listed - but red. I've seen that before where a node would be deleted and show red on the web page. it'd go away after awhile. yet it is still there after 15 hours.

I think there is a network routing issue:
I noticed last night that vmbr1 is not up and should be. vmbr2 has a link and should not. so i must have plugged the patch cable for vmbr1 into vmbr2 . you know sometimes after reinstall eth1 becomes eth2 and eth2 eth1 . or i just screwed up. in any case I'll check fix the wiring..