Re: System alive but can't access from the monitor
On 04/01/2017 11:36 AM, Robin Laing wrote: On 31/03/17 23:28, Paolo Galtieri wrote: I have 2 F25 systems, the problem I'm seeing with the monitor not coming back from sleep only happens on one of them. I did a grep for nouveau on the 2 systems in /var/log/messages. There are no messages including nouveau on the system that does not exhibit the problem. This system has an Intel graphics controller. On this system lspci shows 00:02.0 VGA compatible controller: Intel Corporation Haswell-ULT Integrated Graphics Controller (rev 09) On the one that exhibits the problem I see the following: Mar 31 15:29:31 jackstraw kernel: nouveau :01:00.0: NVIDIA G86 (086100a2) Goes to show that it isn't nouveau related. Now that it shows two different types of video cards, it points to something common I think you misunderstood the message. It is a little confusing, but my understanding of that is that the system that doesn't have the problem has Intel graphics and the one that does have the problem is NVidia. ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: System alive but can't access from the monitor
On 04/04/17 03:54, Tim Jackson wrote: > To add to this thread, I have the exact same problem using > fully-updated (as of today, including libdrm-2.4.76-1) F25, GNOME and > X11 session. To be a bit more specific: for me, when the problem > occurs (after the screen locks and the monitors power down), I *can* > "wake up" the screen, *and* I get a (moveable) mouse cursor, but > behind the cursor is just a static copy of whatever happened to be on > the screen prior to it going to sleep. In most cases, black > (presumably because GNOME blanked the screen prior to lock). Or, if I > happened to be on the login screen when the monitors went to sleep, > possibly a static copy of the login screen. Then your issue is different than mine. And, possibly others. In my case, while xrandr shows both monitors to be "connected" as normal one of the monitors blank and obviously not receiving any signal since it will not stay connected to DP-1 but switches to the HDMI connector which has a Chromecast connected to it. > > As outlined by previous posters, I can't find any way out of this > other than hard rebooting, which is a pretty nasty solution if I had > unsaved stuff. > > On one occasion, the machine apparently fully crashed (not even > pingable), but I can't reproduce that. > > The following kernels are affected: > > kernel-4.10.5-200.fc25.x86_64 > kernel-4.10.6-200.fc25.x86_64 > > If I roll back to kernel-4.9.14-200.fc25.x86_64 the problem goes away. > > lspci -k -nn says: > > 01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GF119 > [GeForce GT 610] [10de:104a] (rev a1) > Subsystem: ASUSTeK Computer Inc. Device [1043:840d] > Kernel driver in use: nouveau > Kernel modules: nouveau > > I have two screens connected via DVI and HDMI respectively. > > To answer Ed Greshko's debugging questions: > >>> 1. Are you using the nouveau driver or the nVidia driver for your >>> card? > > nouveau. In fact this is a pretty fresh install of Fedora all round, > without much extra. > >>> 2. Are you running GNOME under Wayland or X11? > > X11. > >>> If you are using X11, and since you can ssh into the system, what is >>> the output if you do... >>> ssh systemB >>> export DISPLAY=:0 >>> xrandr > > The following (which is exactly the same as under a normal working > session): > > Screen 0: minimum 320 x 200, current 2560 x 1024, maximum 16384 x 16384 > DVI-I-1 connected 1280x1024+1280+0 (normal left inverted right x axis > y axis) 338mm x 270mm >1280x1024 60.02 + 75.02* >1152x864 75.00 >1024x768 75.0360.00 >800x600 75.0060.32 >640x480 75.0059.94 >720x400 70.08 > HDMI-1 connected primary 1280x1024+0+0 (normal left inverted right x > axis y axis) 338mm x 270mm >1280x1024 60.02 + 75.02* >1152x864 75.00 >1024x768 75.0360.00 >800x600 75.0060.32 >640x480 75.0059.94 >720x400 70.08 > >>> Using my system as an example, does doing something like this bring the >>> monitor back? > >> [snip xrandr blah --off, xrandr blah --auto] > > Note that from my description, the monitors are already on and awake, > so they don't really need to be "brought back". But sure, turning them > off and on with xrandr turns them off and on again. It doesn't change > the underlying problem. Right. So, different issues. I don't know if there is a BZ for the issue you're describing. However, it seems what I'm seeing has been around for quite some time. https://bugzilla.redhat.com/show_bug.cgi?id=1179924 and others. -- Fedora Users List - The place to go to get others to do the work for you ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: System alive but can't access from the monitor
On 03/04/17 13:54, Tim Jackson wrote: To add to this thread, I have the exact same problem using fully-updated (as of today, including libdrm-2.4.76-1) F25, GNOME and X11 session. To be a bit more specific: for me, when the problem occurs (after the screen locks and the monitors power down), I *can* "wake up" the screen, *and* I get a (moveable) mouse cursor, but behind the cursor is just a static copy of whatever happened to be on the screen prior to it going to sleep. In most cases, black (presumably because GNOME blanked the screen prior to lock). Or, if I happened to be on the login screen when the monitors went to sleep, possibly a static copy of the login screen. As outlined by previous posters, I can't find any way out of this other than hard rebooting, which is a pretty nasty solution if I had unsaved stuff. On one occasion, the machine apparently fully crashed (not even pingable), but I can't reproduce that. The following kernels are affected: kernel-4.10.5-200.fc25.x86_64 kernel-4.10.6-200.fc25.x86_64 If I roll back to kernel-4.9.14-200.fc25.x86_64 the problem goes away. lspci -k -nn says: 01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GF119 [GeForce GT 610] [10de:104a] (rev a1) Subsystem: ASUSTeK Computer Inc. Device [1043:840d] Kernel driver in use: nouveau Kernel modules: nouveau I have two screens connected via DVI and HDMI respectively. To answer Ed Greshko's debugging questions: 1. Are you using the nouveau driver or the nVidia driver for your card? nouveau. In fact this is a pretty fresh install of Fedora all round, without much extra. 2. Are you running GNOME under Wayland or X11? X11. If you are using X11, and since you can ssh into the system, what is the output if you do... ssh systemB export DISPLAY=:0 xrandr The following (which is exactly the same as under a normal working session): Screen 0: minimum 320 x 200, current 2560 x 1024, maximum 16384 x 16384 DVI-I-1 connected 1280x1024+1280+0 (normal left inverted right x axis y axis) 338mm x 270mm 1280x1024 60.02 + 75.02* 1152x864 75.00 1024x768 75.0360.00 800x600 75.0060.32 640x480 75.0059.94 720x400 70.08 HDMI-1 connected primary 1280x1024+0+0 (normal left inverted right x axis y axis) 338mm x 270mm 1280x1024 60.02 + 75.02* 1152x864 75.00 1024x768 75.0360.00 800x600 75.0060.32 640x480 75.0059.94 720x400 70.08 Using my system as an example, does doing something like this bring the monitor back? [snip xrandr blah --off, xrandr blah --auto] Note that from my description, the monitors are already on and awake, so they don't really need to be "brought back". But sure, turning them off and on with xrandr turns them off and on again. It doesn't change the underlying problem. Is there any known BZ about this? Tim ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org Looks like bugzill 1435000 is the one we are looking at. https://bugzilla.redhat.com/show_bug.cgi?id=1435000 Robin ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
How to view empathy IRC windows?
When I run empathy in F25, all I see is the Contact List which is basically only my AIM contacts. I have used empathy-accounts to add my IRC servers. How do I get empathy to show my IRC servers and allow me to join channels? ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: System alive but can't access from the monitor
To add to this thread, I have the exact same problem using fully-updated (as of today, including libdrm-2.4.76-1) F25, GNOME and X11 session. To be a bit more specific: for me, when the problem occurs (after the screen locks and the monitors power down), I *can* "wake up" the screen, *and* I get a (moveable) mouse cursor, but behind the cursor is just a static copy of whatever happened to be on the screen prior to it going to sleep. In most cases, black (presumably because GNOME blanked the screen prior to lock). Or, if I happened to be on the login screen when the monitors went to sleep, possibly a static copy of the login screen. As outlined by previous posters, I can't find any way out of this other than hard rebooting, which is a pretty nasty solution if I had unsaved stuff. On one occasion, the machine apparently fully crashed (not even pingable), but I can't reproduce that. The following kernels are affected: kernel-4.10.5-200.fc25.x86_64 kernel-4.10.6-200.fc25.x86_64 If I roll back to kernel-4.9.14-200.fc25.x86_64 the problem goes away. lspci -k -nn says: 01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GF119 [GeForce GT 610] [10de:104a] (rev a1) Subsystem: ASUSTeK Computer Inc. Device [1043:840d] Kernel driver in use: nouveau Kernel modules: nouveau I have two screens connected via DVI and HDMI respectively. To answer Ed Greshko's debugging questions: 1. Are you using the nouveau driver or the nVidia driver for your card? nouveau. In fact this is a pretty fresh install of Fedora all round, without much extra. 2. Are you running GNOME under Wayland or X11? X11. If you are using X11, and since you can ssh into the system, what is the output if you do... ssh systemB export DISPLAY=:0 xrandr The following (which is exactly the same as under a normal working session): Screen 0: minimum 320 x 200, current 2560 x 1024, maximum 16384 x 16384 DVI-I-1 connected 1280x1024+1280+0 (normal left inverted right x axis y axis) 338mm x 270mm 1280x1024 60.02 + 75.02* 1152x864 75.00 1024x768 75.0360.00 800x600 75.0060.32 640x480 75.0059.94 720x400 70.08 HDMI-1 connected primary 1280x1024+0+0 (normal left inverted right x axis y axis) 338mm x 270mm 1280x1024 60.02 + 75.02* 1152x864 75.00 1024x768 75.0360.00 800x600 75.0060.32 640x480 75.0059.94 720x400 70.08 Using my system as an example, does doing something like this bring the monitor back? >> [snip xrandr blah --off, xrandr blah --auto] Note that from my description, the monitors are already on and awake, so they don't really need to be "brought back". But sure, turning them off and on with xrandr turns them off and on again. It doesn't change the underlying problem. Is there any known BZ about this? Tim ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: QEMU/KVM/Windows and Samba+NFS
On 04/03/2017 11:47 AM, Tom Horsley wrote: Gah! I meant /etc/nsswitch.conf (too many network related config files I know. But we're also talking about nslookup and ping on Windows. :) ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: QEMU/KVM/Windows and Samba+NFS (SOLVED)
On Mon, 2017-04-03 at 11:29 -0700, Gordon Messmer wrote: > On 04/03/2017 10:49 AM, Patrick O'Callaghan wrote: > > I meant no specific configuration on W. The /etc/hosts file on D has > > existed for a long time with no alteration. > > The /etc/hosts file on D won't help W resolve the name to an IP address. Of course. You asked what I meant by "no configuration" so I was simply explaining that. > > C:\Users\poc>nslookup storage > > ... > > Name: storage > > Address: 192.168.1.65 > > C:\Users\poc>ping 192.168.1.65 > > Reply from 192.168.1.65: ... > > ... > > C:\Users\poc>ping storage > > Ping request could not find host storage. Please check the name and try > > again. > > It's unclear from those snippets whether N actually replied to the ping > requests. Did it? Yes, I didn't transcribe the full reply but it's a proper ping response. > I'm at a loss to explain how nslookup (which is DNS-only) is able to > resolve the name "storage" but ping (which can use DNS and the hosts > file) cannot. That's just bizarre. Isn't it? This makes no sense at all. > > I added storage to that file and rebooted W. No difference. > > Does that mean that ping still can't resolve the name? Or just browsing > for the share? If you still can't "ping storage" then the entry in the > Windows hosts file may not be valid. Send that file, maybe. !! This just in !! I just looked at it again and realised I'd edited the file incorrectly, using the resolv.conf format (name address) rather than the hosts format (address name). Having fixed that, I can now ping storage AND connect to it from the file manager and present credentials as expected. I don't understand how a normal user is supposed to know how to set this up, but (touch wood) things now appear to be working correctly. And I still can't explain the peculiar ping behaviour before I spotted the error, but better let sleeping dogs lie. > (Just so we're clear, we're pursuing two different avenues to resolving > this problem. Adding the name to the Windows hosts file should allow > name resolution, and access through the NAT setup. Setting up bridged > networking should allow direct access and broadcast name resolution.) I haven't gone further with the bridging because of the strange virsh error I reported, but that can wait for now. Many thanks for your patience in helping with this. poc ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Fixed - Re: F24 not recognizing USB drives
Mine is not to reason why Since the last external device I had working was my Samsung Galaxy 4S, I plugged it in. It was recognized. I unmounted it, and then plugged in a USB drive and it was recognized as /dev/sdb. So back in business. With some questions, but if it is working... On 04/03/2017 01:52 PM, Robert Moskowitz wrote: Bunch of suspend/resumes since the last boot. Back home and went to put in a USB drive, and nothing. No /dev/sdb listed. Nothing showing on the desktop as a mountable drive. How would I trouble shoot this? Really don't want to reboot. thanks ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: QEMU/KVM/Windows and Samba+NFS
On Mon, 3 Apr 2017 14:37:24 -0400 Tom Horsley wrote: > A good thing to check when this happens is /etc/resolv.conf on the > system doing the lookups. nslookup just always uses DNS, but other > lookups done by the C library only use the info specified in /etc/resolv.conf. Gah! I meant /etc/nsswitch.conf (too many network related config files :-). ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: QEMU/KVM/Windows and Samba+NFS
On Mon, 3 Apr 2017 11:29:08 -0700 Gordon Messmer wrote: > I'm at a loss to explain how nslookup (which is DNS-only) is able to > resolve the name "storage" but ping (which can use DNS and the hosts > file) cannot. That's just bizarre. A good thing to check when this happens is /etc/resolv.conf on the system doing the lookups. nslookup just always uses DNS, but other lookups done by the C library only use the info specified in /etc/resolv.conf. ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: QEMU/KVM/Windows and Samba+NFS
On 04/03/2017 10:49 AM, Patrick O'Callaghan wrote: I meant no specific configuration on W. The /etc/hosts file on D has existed for a long time with no alteration. The /etc/hosts file on D won't help W resolve the name to an IP address. C:\Users\poc>nslookup storage ... Name: storage Address: 192.168.1.65 C:\Users\poc>ping 192.168.1.65 Reply from 192.168.1.65: ... ... C:\Users\poc>ping storage Ping request could not find host storage. Please check the name and try again. It's unclear from those snippets whether N actually replied to the ping requests. Did it? I'm at a loss to explain how nslookup (which is DNS-only) is able to resolve the name "storage" but ping (which can use DNS and the hosts file) cannot. That's just bizarre. I added storage to that file and rebooted W. No difference. Does that mean that ping still can't resolve the name? Or just browsing for the share? If you still can't "ping storage" then the entry in the Windows hosts file may not be valid. Send that file, maybe. (Just so we're clear, we're pursuing two different avenues to resolving this problem. Adding the name to the Windows hosts file should allow name resolution, and access through the NAT setup. Setting up bridged networking should allow direct access and broadcast name resolution.) ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: QEMU/KVM/Windows and Samba+NFS
On Mon, 2017-04-03 at 10:39 -0700, Samuel Sieb wrote: > On 04/02/2017 09:01 AM, Patrick O'Callaghan wrote: > > This is clunky but works *as long as SElinux is in Permissive mode*. > > Otherwise it complains of permissions. Note that running smbclient on D > > does see the same share (i.e. the NFS-mounted folder) with no issues, > > but of course that's running as me. > > > > What SE context should I apply to the NFS mount to allow me to keep > > SElinux in Enforcing mode but still allow this passthrough to work? > > > > Did you check in the journal for a detailed selinux message? It should > tell you either what the context should be or what boolean you need to set. I'll do that if I can't fix the preferred method of using CIFS directly. poc ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
F24 not recognizing USB drives
Bunch of suspend/resumes since the last boot. Back home and went to put in a USB drive, and nothing. No /dev/sdb listed. Nothing showing on the desktop as a mountable drive. How would I trouble shoot this? Really don't want to reboot. thanks ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: QEMU/KVM/Windows and Samba+NFS
On Mon, 2017-04-03 at 09:25 -0700, Gordon Messmer wrote: > On 04/03/2017 08:14 AM, Patrick O'Callaghan wrote: > > > > It's true that I didn't change the default. However note that W is > > showing STORAGE in its network tab without me doing any specific > > configuration (i.e. this is a new install of Windows 10) so it must be > > getting it from somewhere. > > You later said that you had added "storage" to the hosts file on W, so > I'm not sure whether you did any "specific configuration" or not. I meant no specific configuration on W. The /etc/hosts file on D has existed for a long time with no alteration. > Can you ping "storage" from W? You should be able to. Yes and no: C:\Users\poc>nslookup storage ... Name: storage Address: 192.168.1.65 C:\Users\poc>ping 192.168.1.65 Reply from 192.168.1.65: ... ... C:\Users\poc>ping storage Ping request could not find host storage. Please check the name and try again. > > I assumed it was already using bridging: > > $ route > > 192.168.122.0 0.0.0.0 255.255.255.0 U 0 00 > > virbr0 > > Nope. That's the default NAT networking, not bridged networking. > > > However I tried it anyway (after shutting down W due to excessive caution): > > $ sudo virsh iface-bridge enp3s0 br0 > > error: An error occurred, but the cause is unknown > > Logs might have more information. I'm not sure. I actually haven't > seen that error before. journalctl shows nothing. > The manual setup process for bridged networking is documented here: > https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/7/html/Virtualization_Deployment_and_Administration_Guide/sect-Network_configuration-Bridged_networking.html > > https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/7/html/Networking_Guide/ch-Configure_Network_Bridging.html > > Aside from the initial setup, I think bridged networking is much better > for virtualization. OK, I'll look at that. > > > The other solution would be to edit the W "hosts" file so that it can > > > resolve the name "STORAGE" to an IP address, which *should* also work. > > > > I already have it in /etc/hosts (on W), as "storage" (lower case) but I > > didn't think that mattered with WINS. > > "hosts" on Windows is c:\Windows\System32\Drivers\etc\hosts. I added storage to that file and rebooted W. No difference. poc ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: QEMU/KVM/Windows and Samba+NFS
On 04/02/2017 09:01 AM, Patrick O'Callaghan wrote: This is clunky but works *as long as SElinux is in Permissive mode*. Otherwise it complains of permissions. Note that running smbclient on D does see the same share (i.e. the NFS-mounted folder) with no issues, but of course that's running as me. What SE context should I apply to the NFS mount to allow me to keep SElinux in Enforcing mode but still allow this passthrough to work? Did you check in the journal for a detailed selinux message? It should tell you either what the context should be or what boolean you need to set. ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: QEMU/KVM/Windows and Samba+NFS
On 04/03/2017 08:14 AM, Patrick O'Callaghan wrote: It's true that I didn't change the default. However note that W is showing STORAGE in its network tab without me doing any specific configuration (i.e. this is a new install of Windows 10) so it must be getting it from somewhere. You later said that you had added "storage" to the hosts file on W, so I'm not sure whether you did any "specific configuration" or not. Can you ping "storage" from W? You should be able to. I assumed it was already using bridging: $ route 192.168.122.0 0.0.0.0 255.255.255.0 U 0 00 virbr0 Nope. That's the default NAT networking, not bridged networking. However I tried it anyway (after shutting down W due to excessive caution): $ sudo virsh iface-bridge enp3s0 br0 error: An error occurred, but the cause is unknown Logs might have more information. I'm not sure. I actually haven't seen that error before. The manual setup process for bridged networking is documented here: https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/7/html/Virtualization_Deployment_and_Administration_Guide/sect-Network_configuration-Bridged_networking.html https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/7/html/Networking_Guide/ch-Configure_Network_Bridging.html Aside from the initial setup, I think bridged networking is much better for virtualization. The other solution would be to edit the W "hosts" file so that it can resolve the name "STORAGE" to an IP address, which *should* also work. I already have it in /etc/hosts (on W), as "storage" (lower case) but I didn't think that mattered with WINS. "hosts" on Windows is c:\Windows\System32\Drivers\etc\hosts. Windows typically will not use WINS to resolve network share names any more. ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: QEMU/KVM/Windows and Samba+NFS
On Mon, 2017-04-03 at 07:51 -0700, Gordon Messmer wrote: > On 04/03/2017 03:18 AM, Patrick O'Callaghan wrote: > > Using the Windows file manager on W to open the Network tab, I see > > several entries, including the Samba service on D and the NAS (called > > STORAGE). Opening the D entry shows me the shares from D. Trying to > > open the STORAGE tab gets an error: "Windows cannot access \\STORAGE. > > Check the spelling of the name. Otherwise there might be a problem with > > your network, ...". The log file on N shows no activity. > > > Ah, see... that's not a permission error. That error probably indicates > that W is trying to locate the "STORAGE" host using NMB broadcast, and > can't, which in turn probably means you're using the default NAT > networking under libvirt. It's true that I didn't change the default. However note that W is showing STORAGE in its network tab without me doing any specific configuration (i.e. this is a new install of Windows 10) so it must be getting it from somewhere. > One solution to that problem would be to run "virsh iface-bridge eth0 > br0" to create a bridge interface, and then connect the W VM to the new > bridge. That'll put the guest on the same broadcast domain as D, and > should resolve the problem. I assumed it was already using bridging: $ route Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface default ZyXEL 0.0.0.0 UG 100 0 0 enp3s0 192.168.1.0 0.0.0.0 255.255.255.0 U 100 0 0 enp3s0 192.168.122.0 0.0.0.0 255.255.255.0 U 0 0 0 virbr0 However I tried it anyway (after shutting down W due to excessive caution): $ sudo virsh iface-bridge enp3s0 br0 error: An error occurred, but the cause is unknown > The other solution would be to edit the W "hosts" file so that it can > resolve the name "STORAGE" to an IP address, which *should* also work. I already have it in /etc/hosts (on W), as "storage" (lower case) but I didn't think that mattered with WINS. poc ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: QEMU/KVM/Windows and Samba+NFS
On 04/03/2017 03:18 AM, Patrick O'Callaghan wrote: Using the Windows file manager on W to open the Network tab, I see several entries, including the Samba service on D and the NAS (called STORAGE). Opening the D entry shows me the shares from D. Trying to open the STORAGE tab gets an error: "Windows cannot access \\STORAGE. Check the spelling of the name. Otherwise there might be a problem with your network, ...". The log file on N shows no activity. Ah, see... that's not a permission error. That error probably indicates that W is trying to locate the "STORAGE" host using NMB broadcast, and can't, which in turn probably means you're using the default NAT networking under libvirt. One solution to that problem would be to run "virsh iface-bridge eth0 br0" to create a bridge interface, and then connect the W VM to the new bridge. That'll put the guest on the same broadcast domain as D, and should resolve the problem. The other solution would be to edit the W "hosts" file so that it can resolve the name "STORAGE" to an IP address, which *should* also work. ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: QEMU/KVM/Windows and Samba+NFS
On Mon, 2017-04-03 at 13:26 +0200, Jon Ingason wrote: > Den 2017-04-03 kl. 12:18, skrev Patrick O'Callaghan: > > On Sun, 2017-04-02 at 15:11 -0700, Gordon Messmer wrote: > > > On 04/02/2017 09:01 AM, Patrick O'Callaghan wrote: > > > > I want to use N as a backup server for W, however W cannot see shares > > > > on N. It complains about permissions. > > ... > > > Note that N is configured as a Workgroup server. I'm not using Active > > Directory, though the NAS does support this as an option. Also, W has > > something called Homegroup, apparently Microsoft's attempt to make this > > stuff easier by adding yet another layer of terminology. > > > > poc > > Is your Windows 10 Home or Pro? The reason I ask is when I bought my > last laptop it came with Windows 8 Home and I was unable to mount a > Samba share. I upgraded it then to Windows 8 Pro and could then mount > Samba share. (Now it is upgraded to Windows 10 Pro) It's Home. I could upgrade to Pro if that would make it work. > There are lot of problem with Windows 10 and Samba share but there seams > to be problem with the old SMB 1 and 2 protocol. > > You find more about "HomeGroup" in following URL: > > https://support.microsoft.com/en-us/help/17145/windows-homegroup-from-start-to-finish > I'll look at that, thanks. poc ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: QEMU/KVM/Windows and Samba+NFS
Den 2017-04-03 kl. 12:18, skrev Patrick O'Callaghan: > On Sun, 2017-04-02 at 15:11 -0700, Gordon Messmer wrote: >> On 04/02/2017 09:01 AM, Patrick O'Callaghan wrote: >>> I want to use N as a backup server for W, however W cannot see shares >>> on N. It complains about permissions. ... > Note that N is configured as a Workgroup server. I'm not using Active > Directory, though the NAS does support this as an option. Also, W has > something called Homegroup, apparently Microsoft's attempt to make this > stuff easier by adding yet another layer of terminology. > > poc Is your Windows 10 Home or Pro? The reason I ask is when I bought my last laptop it came with Windows 8 Home and I was unable to mount a Samba share. I upgraded it then to Windows 8 Pro and could then mount Samba share. (Now it is upgraded to Windows 10 Pro) There are lot of problem with Windows 10 and Samba share but there seams to be problem with the old SMB 1 and 2 protocol. You find more about "HomeGroup" in following URL: https://support.microsoft.com/en-us/help/17145/windows-homegroup-from-start-to-finish -- Regards Jon Ingason ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
Re: QEMU/KVM/Windows and Samba+NFS
On Sun, 2017-04-02 at 15:11 -0700, Gordon Messmer wrote: > On 04/02/2017 09:01 AM, Patrick O'Callaghan wrote: > > I want to use N as a backup server for W, however W cannot see shares > > on N. It complains about permissions. > > > It seems to me that from both a performance and reliability perspective, > this is the problem you should solve. Specifically what information are > you given by W when you try to connect to N? You said that N runs > Debian, so I presume you can connect to it via SSH. What information is > logged on N when the client attempts a connection? What version of > Samba is running on N? You're right of course, this is what I should do, but the server doesn't appear to log anything when accessing from Windows, whereas it does when accessing from Linux. In more detail: Using Dolphin (on D) to browse Samba shares I see both the local Samba server on D and the NAS CIFS service on N. Further clicking gets me into the latter with no trouble. The log on N shows this happening. Using the Windows file manager on W to open the Network tab, I see several entries, including the Samba service on D and the NAS (called STORAGE). Opening the D entry shows me the shares from D. Trying to open the STORAGE tab gets an error: "Windows cannot access \\STORAGE. Check the spelling of the name. Otherwise there might be a problem with your network, ...". The log file on N shows no activity. Note that N is configured as a Workgroup server. I'm not using Active Directory, though the NAS does support this as an option. Also, W has something called Homegroup, apparently Microsoft's attempt to make this stuff easier by adding yet another layer of terminology. poc ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org