similar to: [nbdkit PATCH] crypto: Tweak handling of SEND_MORE

This time around, the numbers are indeed looking better than in v1; and I like the interface better. Eric Blake (2): server: Prefer send() over write() server: Group related transmission send()s server/internal.h | 7 +++- server/connections.c | 51 +++++++++++++++++++++++++--- server/crypto.c | 11 ++++--

Slightly RFC, as I need more time to investigate why Unix sockets appeared to degrade with this patch. But as TCP sockets (over loopback to localhost) and TLS sessions (regardless of underlying Unix or TCP) both showed improvements, this looks like a worthwhile series. Eric Blake (2): server: Add support for corking server: Cork around grouped transmission send()s server/internal.h | 3

We disabled Nagle's algorithm to allow less latency in our responses reaching the client; but as a side effect, it leads to more network overhead when we send a reply split across more than one write(). Take advantage of various means for grouping related writes (Linux' MSG_MORE for sockets, gnutls' corking for TLS) to send a larger packet, and adjust callers to pass in our internal

Any time we reply to NBD_CMD_READ or NBD_CMD_BLOCK_STATUS, we end up calling conn->send() more than once. Now that we've disabled Nagle's algorithm, this implies that we try harder to send the small header immediately, rather than batching it with the rest of the payload, which causes more overhead in the amount of actual network traffic. For interfaces that support corking (gnutls, or

On Fri, Mar 24, 2023 at 11:47:20AM +0100, Florian Westphal wrote: > qemu-nbd doesn't set TCP_NODELAY on the tcp socket. > > Kernel waits for more data and avoids transmission of small packets. > Without TLS this is barely noticeable, but with TLS this really shows. > > Booting a VM via qemu-nbd on localhost (with tls) takes more than > 2 minutes on my system. tcpdump

As mentioned in the previous patch, sending small packets as soon as possible leads to network packet overhead. Grouping related writes under corking appears to help everything but unencrypted Unix plain sockets. I tested with appropriate combinations from: $ nbdkit {-p 10810,-U -} \ {--tls=require --tls-verify-peer --tls-psk=./keys.psk,} memory size=64m \ --run

v1 was here: https://www.redhat.com/archives/libguestfs/2020-February/msg00081.html v2 replaces struct connection *conn = GET_CONN; with GET_CONN; which sets conn implicitly and asserts that it is non-NULL. If we actually want to test if conn is non-NULL or behave differently, then you must use threadlocal_get_conn() instead, and some existing uses do that. Rich.

The third patch is a large but mechanical change which gets rid of passing around struct connection * entirely within the server, preferring instead to reference the connection through thread-local storage. I hope this is a gateway to simplifying other parts of the code. Rich.

Since commit 86fdb48c6a5362d66865493d9d2172166f99722e we have stored the connection object in thread-local storage. In this very large, but mostly mechanical change we stop passing the connection pointer around everywhere, and instead use the value stored in thread-local storage. This assumes a 1-1 mapping between the connection and the current thread which is true in *most* places.

libnbd's copy/copy-nbd-error.sh was triggering an assertion failure in nbdkit: $ nbdcopy -- [ nbdkit --exit-with-parent -v --filter=error pattern 5M error-pread-rate=0.5 ] null: ... nbdkit: pattern.2: debug: error-inject: pread count=262144 offset=4718592 nbdkit: pattern.2: debug: pattern: pread count=262144 offset=4718592 nbdkit: pattern.1: debug: error-inject: pread count=262144

> As a streaming server, it's fairly crucial for icecast to > send out data with as low a delay as possible (many clients > don't care, but some do). That's why we use TCP_NODELAY - we > actually WANT to send out data as soon as we can. Nagle is inherently unsuited for streams. NODELAY was (imho) ment for connections for which Nagle isn't sufficient and CORK is not

I did chmod -R 777 /var/lib/dovecot and it worked. I just need to figure out what the correct setting are. Thanks everyone //Ger ----------------original message----------------- From:Ger Hooton Scoil Chro? ?osa Blarney [ghooton at scins.ie ] To:Robert Moskowitz [rgm at htt-consult.com ], chaouche yacine [yacinechaouche at yahoo.com ], dovecot at dovecot.org Date:Tue, 21 Mar 2017 14:05:17

dovecot-v1.0-alpha build failed for Solaris 11 (OpenSolaris Nevada). The problematic line is in socket.c line 228. The fix should be to change SOL_TCP to IPPROTO_TCP found in netinet/in.h. This change should work universally on all platforms. Gary

On Mon, 5 Jul 2021 at 07:15, Hooton, Gerard <g.hooton at ucc.ie> wrote: > > Hi All, > I am having problems with a kickstart install of CentOS 8 > When I try to do a completely automated install using PXE/UEFI it get to the point where it reads the kickstart config file. > Then I see the following message > "kickstart install Started cancel waiting for multipath

Hi everyone Has anybody ever come across this, I want it to join automatically the recording files for me but it creates the wav files as I expect but it doesn't join them sadly, so eg: ls 1250623104.10-in.wav 1250628672.23-in.wav 1250630411.30-in.wav 1250623104.10-out.wav 1250628672.23-out.wav 1250630411.30-out.wav Here is my queues.conf snippet [testQ]

Hello all, I am using Samba version 3.2.5 configured for Windows Domain login. It was working fine until I moved over to ldapsam. Now when I login to the XP client I don't see the logon drive (X: in my case) Any debug tips? //Ger -- Gerard Hooton Systems Administrator Blarney Boy's National School, Blarney. Co. Cork

Hi All, I cannot get the sieve_before to work. I have dovecot & sieve installed on Debian 7 as follows /etc/dovecot/conf.d/90-sieve.conf : plugin { # The path to the user's main active script. If ManageSieve is used, this the # location of the symbolic link controlled by ManageSieve. sieve = ~/.dovecot.sieve # The default Sieve script when the user has none. This is a path to a

The following is a message to be sent to the President of the United States of America. Although we may not be able to do a great deal from where we are, but for the people of America just knowing we care and feel their sadness will help. Please put your name on the following list and send it to all you know and who care. If you are the 100th name and every 100th there on could you please also

The NBD spec says that additional information in response to NBD_OPT_GO or NBD_OPT_INFO, such as NBD_INFO_NAME, is optional, but also recommends that if a client hints about wanting a particular piece of information, that it is better for the server to reply to those known hints. This is also the the only way for a client to learn the canonical name that is used when connecting to the default

Hi everybody I have a logic question that is confusing me. ifTime(00:00-12:00|*|*|*) { Playback(welcome-morning); } else { ifTime(12:00-18:00|*|*|*) { Playback(welcome-afternoon); } else { Playback(welcome-evening); }