Posts made by kurtblah

I’ve set the FOG_MEMTEST_KERNEL in the General Settings to memtest.bin with no luck, same error. There is a file called “memdisk” in the same folder, it errors in the same way.

Yes. the Memtest ISO boots fine from a USB drive via UEFI.

I am wondering, are other users having this same problem with the FOG installed Memtest when booting as UEFI clients? I’d like to know if my environment here is playing a role here.
Are users with UEFI clients needing to set their PXE menu exit to REFIND_EFI in order to boot to the disk from the FOG menu? For me its the only way to get the our machines into the OS from the boot menu.

Nope, its off.

@george1421 The new 7.0 memtest ISO does not have those two files (memtest and memtest.bin) within it. The 7.0 version is a different beast it seems.

@george1421 memtest 5.01 comes up from the BIOS boot menu. Memtest site latest version is 7.0.

http://www.memtest86.com/technical.htm#network

Perhaps their “Site Edition” is needed for this?

http://www.memtest86.com/features.htm

With REFIND_EFI as the exit type I can boot to the hard diskand successfully run the other options, but when the MEMTEST option is chosen I get the chainloading error. The exit type does not have an effect on the Memtest behavior, it gives the same error regardless of the exit type set in the PXE Boot Menu settings within the web gui.

Sorry, should have mentioned that I am using 1.3.0 RC1.

UEFI clients need to be set with the REFIND_EFI exit option to boot to the hard disk.

From the FOG menu I can run Memtest from the legacy BIOS boot clients, but it fails to start from UEFI clients’ FOG menu.

2e008081
exec format error
chainloading failed

Any idea what is causing this?

thanks, -kurt

@Sebastian-Roth I had to reinstall node.js, but I followed a different set of directions than before, and wouldn’t you know, node-dhcproxy functions as it should. I’m guessing that I did not have node.js setup properly the last time I ran it. I’ve attached the dump file anyway.

0_1469053427459_dump.zip

Thanks for your help on this, it is nice to have another option for our network environment should dnsmasq becomes unusable. -kurt

@Sebastian-Roth yes, consistent behavior every time, and almost instantly when a the client starts up. That’s about all the info I can give you, along with the screen capture above. I quite liked the one from the CloneDeploy project and its debug mode, too bad it only came in the windows flavor. I also tried out PyPXE and proxyDHCPd.

That did it, problem solved. Thanks for the info on the new DNSMasq binary.

-kurt

I’m seeking a single-box solution for our fog servers used in our labs. Legacy BIOS is currently working great with the trunk builds, but I am trying to sort out a solution to support both BIOS and UEFI systems. Running DNSMasq on the fog server supports BIOS machines fine, but, as noted in previous threads, UEFI and BIOS cannot coexist with the version of DNSMasq I am using (on Fedora 23).

The windows-based CloneDeploy proxy DHCP server, when run on windows client connected to the LAN, works as it should, supporting BIOS and UEFI machines that boot from the FOG server’s /tftpboot folder. However, I’d rather not have to worry about keeping track of this task on different hardware.

My question is, is there something similar to this to run on the fog server instead of DNSMasq? Or does the most recent DNSMasq version solve the UEFI issue?

I have tried using node-dhcproxy but it bombs out when a machine is discovered:

[root@localhost node-dhcproxy]# node dhcproxy.js
server listening: 0.0.0.0:67
proxy listening: 0.0.0.0:4011
14 Jul 13:55:14 - Got DHCPDISCOVER from 0.0.0.0:68 (54:04:a6:2f:3d:1e)  with packet length of 548 bytes
Received DHCP DISCOVER
buffer.js:69
      throw new Error(
      ^

Error: If encoding is specified then the first argument must be a string
    at new Buffer (buffer.js:69:13)
    at Packet.toBuffer (/root/git/node-dhcproxy/lib/packet.js:58:12)
    at DHCPServer._send (/root/git/node-dhcproxy/lib/server.js:47:19)
    at DHCPServer.offer (/root/git/node-dhcproxy/lib/server.js:75:15)
    at dhcpd.discover (/root/git/node-dhcproxy/lib/dhcpd.js:97:17)
    at emitTwo (events.js:106:13)
    at DHCPServer.emit (events.js:191:7)
    at DHCPServer._emitPacket (/root/git/node-dhcproxy/lib/server.js:60:15)
    at DHCPServer.<anonymous> (/root/git/node-dhcproxy/lib/server.js:24:20)
    at emitTwo (events.js:106:13)
[root@localhost node-dhcproxy]# 

Any suggestions or direction on approaching this would be greatly appreciated. Solving this ProxyDHCP issue is the last hurdle in my path towards simplified lab support ( I am a college art professor with no IT staff support…)

Thanks, -kurt

Disabling IPv6 did the trick. I ran:

 sudo echo "1" > /proc/sys/net/ipv6/conf/all/disable_ipv6

and the installer ran normally. I’ll need to shut it down permanently.

Thank you for your help. Is this IPv6 issue consistent in all of the builds?

-kurt

no problem. here’s the ip addr:

[root@localhost bin]# ip addr show
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: enp1s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
    link/ether d4:85:64:01:6a:05 brd ff:ff:ff:ff:ff:ff
    inet 192.168.1.251/24 brd 192.168.1.255 scope global enp1s0
       valid_lft forever preferred_lft forever
    inet6 2601:243:200:d722::73a8/128 scope global dynamic
       valid_lft 592005sec preferred_lft 592005sec
    inet6 2601:243:200:d722:d685:64ff:fe01:6a05/64 scope global noprefixroute dynamic
       valid_lft 226861sec preferred_lft 226861sec
    inet6 fe80::d685:64ff:fe01:6a05/64 scope link
       valid_lft forever preferred_lft forever

[root@localhost bin]# systemctl restart httpd
Job for httpd.service failed because the control process exited with error code. See "systemctl status httpd.service" and "journalctl -xe" for details.

[root@localhost bin]# systemctl status httpd -l
● httpd.service - The Apache HTTP Server
   Loaded: loaded (/usr/lib/systemd/system/httpd.service; enabled; vendor preset: disabled)
   Active: failed (Result: exit-code) since Mon 2016-06-20 19:23:42 CDT; 15s ago
  Process: 2975 ExecStart=/usr/sbin/httpd $OPTIONS -DFOREGROUND (code=exited, status=1/FAILURE)
 Main PID: 2975 (code=exited, status=1/FAILURE)

Jun 20 19:23:42 localhost.localdomain systemd[1]: Starting The Apache HTTP Server...
Jun 20 19:23:42 localhost.localdomain httpd[2975]: AH00526: Syntax error on line 4 of /etc/httpd/conf.d/fog.conf:
Jun 20 19:23:42 localhost.localdomain httpd[2975]: Invalid command '2601:243:200:d722::73a8', perhaps misspelled or defined by a module not included in the server configuration
Jun 20 19:23:42 localhost.localdomain systemd[1]: httpd.service: Main process exited, code=exited, status=1/FAILURE
Jun 20 19:23:42 localhost.localdomain systemd[1]: Failed to start The Apache HTTP Server.
Jun 20 19:23:42 localhost.localdomain systemd[1]: httpd.service: Unit entered failed state.
Jun 20 19:23:42 localhost.localdomain systemd[1]: httpd.service: Failed with result 'exit-code'.```

-kurt

This is where the installer bombs out:

 * Setting up SSL FOG Server...................................OK
 * Restarting Apache2 for fog vhost............................Failed!

From what I can tell, the installer is adding lines to the fog.conf file that keeps the httpd service from restarting.

I can take out the bad lines from the fog.conf file and the httpd service restarts as it should.

-kurt

systemctl enable mariadb

returns nothing. running the mysql command works fine.

[root@localhost bin]# mysql
Welcome to the MariaDB monitor.  Commands end with ; or \g.
Your MariaDB connection id is 3
Server version: 10.0.23-MariaDB MariaDB Server

Copyright (c) 2000, 2015, Oracle, MariaDB Corporation Ab and others.

Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.

MariaDB [(none)]> \q
Bye

The munged lines in the fog.conf vhost file and the installer output look to me like IPv6 addresses

@Wayne-Workman

[root@localhost ~]# systemctl enable mysqld
Failed to execute operation: No such file or directory
[root@localhost ~]# systemctl enable mysql
Failed to execute operation: No such file or directory

MariaDB is installed and works.