On Sun, Oct 28, 2018 at 11:41:14AM +0100, Bernhard Übelacker wrote: > Dear Maintainer, > just some more information, because I think I see this > difference in my qemu buster amd64 VM too. > > Before I could ssh into that machine just after some seconds. > > Now it takes some time until that line "random: crng init done" > appears in dmesg. > With logging in in the qemu window this line appears just after a > few seconds, when just trying via ssh it takes much longer. I have a NAS running Debian[1] and there I only get the "crng init done" message after more than 15 minutes (even though there is network activity and I debugged the failure via rs232): root@machine:~# dmesg | tail [ 22.579884] audit: type=1400 audit(1529665920.708:3): apparmor="STATUS" operation="profile_load" profile="unconfined" name="man_filter" pid=299 comm="apparmor_parser" [ 22.627877] audit: type=1400 audit(1529665920.708:4): apparmor="STATUS" operation="profile_load" profile="unconfined" name="man_groff" pid=299 comm="apparmor_parser" [ 22.711102] 8021q: 802.1Q VLAN Support v1.8 [ 22.840150] mvneta d0074000.ethernet eth1: PHY [d0072004.mdio-mii:01] driver [Marvell 88E1318S] [ 22.856020] mvneta d0074000.ethernet eth1: configuring for phy/rgmii-id link mode [ 22.872501] IPv6: ADDRCONF(NETDEV_UP): eth1: link is not ready [ 26.020202] mvneta d0074000.ethernet eth1: Link is Up - 1Gbps/Full - flow control rx/tx [ 26.027639] IPv6: ADDRCONF(NETDEV_CHANGE): eth1: link becomes ready [ 1057.771583] random: crng init done [ 1057.774739] random: 7 urandom warning(s) missed due to ratelimiting I'm not aware the machine has a random number generator, so the solutions presented here up to now don't help me. I think the idea to mention this in the release notes for buster is a good one (unless a solution is found until then of course). Best regards Uwe [1] this i an armv7/mvebu machine with a Armada 370 CPU
Attachment:
signature.asc
Description: PGP signature